A two-dimensional model of glacier flow is presented which includes periodical surging as a natural phenomenon for a certain class of glaciers. The input consists of the bedrock and balance profiles along the glacier, together with the ice flow properties and a frictional lubrication factor. The basal stress is determined from the condition of gross equilibrium for the whole glacier, together with the distribution of the frictional lubrication from energy dissipation along the glacier.

The difference between the basal stress and the down-slope stress of the glaciers produces longitudinal strain-rates which determine the basal sliding velocity. Since the velocity is also involved in the frictional lubrication, feed-back develops between the basal stress and sliding velocity.

For a given lubrication factor, a critical stage can be reached for which the velocity becomes sufficiently high to lower the basal stress, enough to cause very high velocities to develop. The model thus gives rise to three classes of glaciers with two modes of flow.

“Ordinary” glaciers do not have sufficient mass flux, for the given bedrock profile, to go beyond the “slow mode” in which the basal stress and velocity increase together as the glacier builds up to steady state.

“Fast” glaciers have sufficient flux to remain continuously in the “fast mode” with high velocities and relatively low basal stress.

“Surging” glaciers have sufficient flux to reach the fast mode but not sufficient to maintain it, and thus develop a periodically oscillating state between the fast and slow modes with gradual build up and rapid drainage.

Sample results are presented for models of a typical large valley surging glacier and for a very high-speed surging glacier.

Microscopic and textural observations were made on ice samples cored from Blue Glacier slightly below the equilibrium line to depths of 60 m. Observations were started within a few minutes after collection. Water was found in veins along three-grain intersections, in lenses on grain boundaries and in irregular shapes. Gas was found in bubbles in the interior of crystals, in bubbles touching veins, and locally in veins. Vein sizes showed some spread; average cross-sectional area was about 7 × 10−4 mm2 with no discernible, trend with texture or depth except within 7 m of the surface. Before the samples were examined they could have experienced a complex relaxation which could have changed them significantly. As a result it is not possible to determine the in situ size of veins, but an upper limit can be determined. Also it is not possible to predict intergranular water flux per unit area, but 1 × 10−1 m a−1 represents an upper limit. In coarse-grained ice the water flux density is likely to be even smaller, because of a low density of veins, and blocking by bubbles. This indicates that only a very small fraction of the melt-water production on a typical summer day can penetrate into the glacier on an intergranular scale except possibly near the surface. The existence of conduit-like features in several cores suggests that much melt water can nevertheless penetrate the ice locally without large-scale lateral movements along the glacier surface. The observed profile of ice temperature indicates that the intergranular water flux may be much smaller than the upper limit determined from the core samples.

Studies that had been started in 1971 were continued in the summer of 1972 in order to gather information on the intraglacial run-off system in the ablation region of Hintereisferner, In a long-term test of 30 h, the daily fluctuations of the inflow into a moulin and the discharge at the glacier mouth were measured. A number of injections of dye tracers were made to measure the curve of concentration versus time. Finally, a test was made to compare with last year's results, in which the inflow was marked with dye at three different points on the glacier. From the long-term measurements, the daily fluctuations of the average velocity of flow in the intraglacial run-off system and the dispersion coefficient were determined. The results confirm the experiments of last year, showing that the melt water reaches the glacier bed taking the shortest possible route without passing through cavities within the glacier, flowing mostly in an open channel.

Temperatures have been measured at two sites in a temperate glacier (Blue Glacier, Washington, U.S.A.) to depths of 192 and 76 m. The accuracy, which varies between about 0.002 and 0.005 deg, is about an order of magnitude better than previously obtained. Except near the surface, temperatures vary linearly with depth but are in disagreement with the simplest model of a temperate glacier, being about 0.02 deg colder near the surface and 0.04 deg colder at 192 m depth.

A deep-ice temperature measurement program has been conducted on Trapridge Glacier, Yukon Territory. Large regions of temperate ice are predicted at the base of the otherwise cold glacier, The glacier snout, frozen to bedrock, appears to act as an ice dam allowing the build up of an ice reservoir in the upper regions. Thermal regulation of the surges of Trapridge Glacier is suggested and the relevance of basal temperatures in large surging glaciers is discussed.

Boudinage structures have only rarely been reported in glacier ice, yet they seem to be widespread in Swiss glaciers. They form in debris-free, strongly foliated ice by the stretching, necking and rupture of layers or groups of layers, when the principal compressive strain axis lies at a high angle to the layering. Two main types of boudinage are distinguished. The first results from the difference in competence between fine-grained and coarse-grained ice, and indicates that the former is more resistant to flow than the latter. The second occurs in more equigranular ice which shows a strong planar anisotropy; associated with the necking of such ice is the development of shear planes, along which the layers are displaced. As in deformed rocks, it is not possible to determine the directions of the finite principal strain axes from the boudinage structures alone. Although the boudins described here all occur in longitudinal foliation, it is suggested that they are likely to form in other situations also.

Internal reflections are due to changes in electrical admittance between adjacent depositional layers. Reflection coefficients are given for discontinuous changes in either the permittivity or loss tangent. The observed strengths of internal echoes rule out the possibility that they are caused by isolated layers containing “foreign” material, but suggest instead that they are due to systematic fluctuations of density, anisotropy, or loss tangent. The electrical behaviour of ice from polar ice sheets is reviewed and compared with that of ice grown in controlled laboratory conditions. We suggest that the impurity distribution in polycrystalline ice is dependent on the impurity content and the temperature of freezing, and the conductivity is essentially determined by the intrinsic and impurity defects within the crystal lattice. In a polar ice sheet, density fluctuations decrease with depth, whereas loss tangents (and hence their fluctuations) increase since the ice becomes warmer towards bedrock. Echo strengths in central Antarctica are compared with those calculated for a boundary where either all bubbles disappear or the loss tangent changes by 50%. Assuming a constant layering geometry to 2 700 m depth, density fluctuations account for echoes above 1 500 m, but deeper echoes are best explained by variations in the ice conductivity.

In Svalbard and Iceland there appears to be much more debris entrained in glaciers that surge than in those which do not. Conditions particularly favourable for the basal incorporation of debris develop as a consequence of the high flow velocities attained by a surge. These are increased cavitation in the lee of obstacles and an increased supply of basal melt water resulting from frictional heat and from the trigger zone. Layers of regelation ice incorporating debris can thus develop to a much greater vertical and horizontal extent than in non-surging glaciers. Excessive shearing, and the distortion of foliation structures in the terminal zone of compressive flow, enhance the vertical development of the debris-rich regelation layers. Glaciers that surge over outwash and/or fjord-bottom sediments become particularly rich in debris.

The surface-perpendicular component of velocity and strain-rate have been determined at one site in the ablation area of Blue Glacier, Washington, U.S.A., where the total depth is about 250 m. The strain-rate is near zero at the surface but increases with depth to about 4% a−1 at 175 m. The results were obtained with the help of a finite deformation theory from the measured stretch of cables frozen into the ice.

A frozen mud layer plastered at the base of the Glacier d’Argentière in the Western Alps acts as an ion-exchange membrane for the interstitial water passing through it, causing electrolyte filtration and relative retardation of cations. This interstitial water, which is forced under a pressure gradient of about 3 bar cm−1, refreezes in the form of ice accretions.

A comparison of the chemical composition of these ice accretions and water held at their surface (corresponding to the situation after the forcing) and the water in contact with the mud layer (corresponding to the situation before the forcing) shows a desalinating effect, a monovalent-divalent cation species separation, and a sorting effect between the alkalies as well as between the alkaline-earths. This phenomenon has to be taken into account in the study of the chemical composition of stratified ice containing interbedded mud layers which occurs in the basal part of several glaciers.

This study was undertaken to determine the sources of debris and methods of transportation and deposition in and on a small cirque glacier. Data were collected on the amount of debris, stone roundness, the presence of striations and polish, and till fabric. Lichenometry gave relative ages of the tills, and suggests that the Gannett Peak till is of at least three ages and probably overlies Audubon till.

Debris originating from subglacial erosion can be differentiated from that from rockfall or avalanches on stone roundness, polish and striations. A maximum of 70% of the present glacial load derives from subglacial erosion, as compared to 88% during the Gannett Peak stade. Rockfall rates are 35-50 m3/year at present and were 290–485 m3/year during the Gannett Peak stade. Data on present-day processes and on the volume and age of Gannett Peak moraines can be used to make comparisons on present and past rates of denudation. The denudation rate in the cirque at present is 95–165 mm/1 000 year; in the past it was 4 920-8 160 mm/ 1 000 year. The denudation rate and the glacial effects on debris are comparable to rates from other glacial areas and effects on debris carried by valley glaciers and continental ice sheets.

Compressional wave velocity gradients at 43 of 50 Antarctic traverse stations plot as sequences of straight lines on semilogarithmic graph paper. Intersections of the lines appear to correlate with depths at which the predominant metamorphic mechanism in polar snow changes. The seismie pattern supports a three-layer interpretation of snow densification. The base of the upper layer (8.4±12.3 m) corresponds to the “critical depth” of Anderson and Benson (1963) at which snow grains settle into a “random closepacked” arrangement. The base of the lower layer may correspond to the firn: ice transition depth, but more data are needed to confirm this conclusion. It is unclear what densification phenomenon is marked by the base of the middle layer (27.7±4.4 m). The distinction between the middle and lower layers tends to disappear and the velocity gradient at a fixed depth increases as mean annual accumulation decreases.

Equilibrium-line altitudes on the White Glacier, Axel Heiberg Island, and the north-west sector of the Devon Ice Cap are shown to be closely related to mean July freezing-level heights at nearby upper-air weather stations. An inverse relationship between July freezing-level heights and mass balance on the Devon Ice Cap is also shown. Reasons for such correlations are suggested and some limitations of the relationship are outlined. Recent lowering of the freezing level in July is discussed in relation to the theoretical “steady-state” equilibrium-line altitudes in the Canadian high Arctic. It is suggested that positive mass-balance years have predominated over a large part of northern Ellesmere Island and north-central Axel Heiberg Island since 1963, and some glaciological evidence supporting this hypothesis is given.

The displacement of the surface of an ice sheet and of markers set in its top layers can be measured geodetically, and also, it is expected, by radio-echo methods. The paper discusses how such measurements could be interpreted as showing long-term changes in the thickness of the ice sheet; in particular it discusses how one might design an experiment so as to avoid unwanted effects due to short-term changes in rate of accumulation. The analysis is similar to that of Federer and others (1970), but it corrects an error, so that when applied to their results for central Greenland it gives a different result for the lowering of the surface. Federer and others have already concluded that the average accumulation rates during the past 100 years have been below those needed to keep in balance with the velocity of the ice sheet as a whole. Using a particular model, it is found that this has resulted in the surface lowering at a mean rate of 0.050 m a−1 between 1871 and 1968, and a mean rate of 0.140 m a−1 between 1959 and 1968.

An approximate analysis of (he processes of inciting and freezing of a drill hole, 500 m depth and 0.15 m m initial radius, through an ice shelf was made. Results are expressed in graphical form showing the time available for experimentation under the hole as a function of heating duration. It was also found that refreezing has a much slower rate than melting.

Study of the waning snow-pack along the crest of the White Mountains of California and Nevada in 1970 and 1974 indicates that a significant proportion of the high-altitude snow in the range sublimates and/or evaporates shortly after melting. Qualitative and limited quantitative evidence suggest that the amount of snow thus disposed of may be as much as 50-80% of the total springtime pack. Meteorological observations in the White Mountains demonstrate that atmospheric conditions particularly favorable for effective sublimation/evaporation are common in May and June, the main ablation period in the mountains. The general lack of evidence of surficial erosion on slopes above 3 500 m, often supposed to be wide-spread and caused by so-called "snow-melt" runoff, is therefore readily explained-there is little "snow-melt" erosion simply because there is only limited snow-melt.

A high-resolution radio echo sounder operating at a frequency of 620 MHz has been developed for studies of temperate glaciers. Excellent spatial resolution is obtained through the use of a short pulse length (70 ns) and an antenna beam width of 5.2°. Large amounts of high-quality data may be rapidly collected since the sounder incorporates an automatic positioning system and an on-line computer. Real time analysis of the echoes facilitates the understanding of complex reflecting horizons observed in temperate glaciers.

Results obtained during field trials of the echo sounder on both the Wapta Icefield and Athabasca Glacier are given. Intraglacial structures which may be due to water levels within the ice have been detected.

The concentration of microparticles in the 2 164 m long ice core from “Byrd” station Antarctica, varies cyclically. Highest concentrations of 0.65 μm diameter microparticles occur where oxygen-isotope studies show lowest paleotemperatures. The age of the bottom ice estimated from microparticle-concentration variations, assuming an annual cycle, is 27 000 years, much less than from oxygen-isotope studies.

The study of ice in the upper Great Lakes, both from the operational and the scientific points of view, is receiving continued attention. Quantitative and qualitative field work is being conducted to provide the needed background for accurate interpretation of remotely sensed data. The data under discussion in this paper were obtained by a side-looking multiplexed airborne radar (SLAR) supplemented with ground-truth data.

Because of its ability to penetrate adverse weather, radar is an especially important instrument for monitoring ice in the upper Great Lakes. It has previously been shown that imaging radars can provide maps of ice cover in these areas. However, questions concerning both the nature of the surfaces reflecting radar energy and the interpretation of the radar imagery continually arise.

Our analysis of ice in Whitefish Bay (Lake Superior) indicates that the combination of the ice/water interlace and the ice/air interface is the major contributor to the radar backscatter as seen on the imagery. At these frequencies the ice has a very low relative dielectric permittivity (< 3.0) and a low loss tangent. Thus, this ice is somewhat transparent to the energy used by the imaging SLAR system. The ice types studied include newly formed black ice, pancake ice, and frozen and consolidated pack and brash ice.

Although ice thickness cannot be measured directly from the received signals, it is suspected that by combining the information pertaining to radar backscatter with data on the meteorological and sea-state history of the area, together with some basic ground truth, better estimates of the ice thickness may be provided. In addition, certain ice features (e.g. ridges, ice-foot formation, areas of brash ice) may be identified with reasonable confidence. There is a continued need for additional ground work to verify the validity of imaging radars for these types of interpretations.

Laboratory experiments show that pieces of melting ice about one centimetre across, moving under normal loads across roughened glass surfaces, travel much faster than regelation theory predicts. The discrepancy is as much as 40 times for the finest scale surfaces, (prepared by grinding with carborundum particles of 60 μm mean diameter) and increases further if the load is reduced below three bars. On the other hand melting ice moves, under similar conditions, across rough porous glass surfaces at approximately the speed predicted by regelation theory. We suggest the reason is that melt water, produced by the dissipation of frictional energy, accumulates at the interface between ice and ground glass where it promotes sliding, but can easily drain away from a porous surface. Similar effects at the bed of a temperate glacier may cause the contribution of regelation to the bed-slip process to depend sensitively upon the melt-water regime.