The physics of water flow within and under glacier ice is examined with special reference to the periodic catastrophic outbursts of water (jökulhlaups) from the subglacial lake Grímsvötn, Vatnajökull, Iceland. The lake is sealed until it reaches a critical level which enables it to lift the glacier, helped by a hydrostatic cantilever effect. The differential equations for non-steady water flow in a subglacial tunnel are derived and applied to the 1972 Grímsvötn outburst. The discharge: time relation observed during the growth stage, and the abrupt ending of the flood, are both very well accounted for by a theory which is insensitive to the details of the subglacial tunnel system. The steady state, in which an intergranular vein or tunnel is simultaneously melted open by frictional heat and closed by plastic deformation, may be stable or unstable according to the conditions imposed at the ends. This explains why the flow of water in a vein does not normally increase unstably as in a jökulhlaup. An ice-dammed lake does not drain away through the vein system because the driving force on the vein-water is towards the lake rather than away from it.

The glacier sliding theory of Nye is modified to include the effect of solutes in subglacial regelation waters on the sliding process. Motivation for this development stems from studies of subglacially formed chemical deposits that appear to be widespread on rock surfaces recently exposed by retreating temperate glaciers. These deposits indicate clearly that considerable local concentration of solutes commonly occurs subglacially as a result of the selective rejection of solutes into the melt during the freezing associated with regelation sliding. Because solutes accumulate where regelation waters refreeze, they tend to lower the temperature there and hence inhibit the heat transport away from these areas that is essential for regelation sliding. For a simple sinusoidal bed and solute distribution in the regelation water film, the modified theory shows that a maximum excess of solutes of, for example, several millimoles/1 of dissolved CaCO3 along lee surfaces relative to stoss surfaces impedes basal sliding significantly, especially if the bed roughness wavelength does not exceed about one meter.

The elastic stress intensity factor is a parameter used in fracture mechanics to describe stress conditions in the vicinity of the tip of a sharp crack. By superimposing solutions of stress intensity factors for different loading conditions, equations are derived which model crevasses in ice. Solutions are presented for the theoretical depth of isolated crevasses, free from or partially filled with water. Close agreement exists with a previous calculation by Weertman using a different technique. The effect of crevasse spacing is investigated and it is demonstrated that closer spacing always reduces crevasse depth, but over a wide range of spacing the predicted variation in depth is slight.

This paper contains a critical survey of the literature on the bearing capacity of floating ice plates. It consists of a discussion of general questions, a critical survey of analytical attempts to determine the bearing capacity of floating ice plates and a survey of field and laboratory tests on floating ice plates and their relation to the analytical results. It concludes with a systematic summary of the results, a discussion of observed shortcomings, and suggestions for needed investigations.

Three experiments were conducted to assess acoustic properties of hard-pack snow. One test involved transmission of acoustic signals in the frequency range 100-20 000 Hz through natural snow-pack in order to measure signal loss of a point acoustic source. At all frequencies the relatively high-energy input signal decays rapidly by energy dissipation, with nominal diffusion occurring at large distances from the source. Signal persistence is greatest in the frequency range 100-200 Hz, In a second test, acoustic bursts in snow columns under deformation were recorded. Spectrum analysis in the frequency range 500-14000 Hz reveals dominance of signal amplitudes at frequencies between 1 000 and 10 000 Hz. This dominance is attributed to the strong attenuation properties of snow and suggests the use of waveguide or collector techniques to monitor natural acoustic emissions in snow-pack. In a third test several waveguide geometries and materials were evaluated for their acoustic signal interception and transmission characteristics. In general, metallic waveguides show the least attenuation of the configurations tested.

In this paper a continuum mechanical formulation of snow pressure is given. The snow pressure against retaining structures on slopes is considered in two separate parts: (1) the static component due to the regression of transverse expansion as discussed by Haefeli, and (2) the dynamic component due to the interruption of the creep (internal deformation) and glide (slip of the entire snow cover over the ground). Snow-pressure calculations made using the non-linear viscous constitutive equations given by McClung (in press) are given for the plane strain-rate problem of a rigid barrier on a snow-covered slope. These calculations are compared with the previous formulation of snow pressure given by Haefeli for the expected range of boundary conditions for the structure and for the snow-earth interface. The results show that the original formulation by Haefeli gives a dynamic component of similar magnitude to the present calculations. Substantial differences are apparent when the up-slope distances for interruption of the creep and glide processes are compared.

Studies of fluted surfaces beyond the margins of glaciers in Spitsbergen, Iceland, Norway and the Alps show that almost all emanate from rigid obstructions, commonly boulders in till. Field relations of flutes are described and it is shown that a relationship exists between flute height and the height of the initiating obstruction. Subglacial observations indicate that flutes form when till is intruded into tunnels which tend to open up in the lee of obstacles. The pattern of strain implied by this process is shown to be reflected by micro- and macrofabrics in the till. The commonly found occurrence of an average spacing between flutes does not arise because of some rhythmic or periodic mechanism in the glacier, but is produced by the random seeding of boulders which themselves generate flutes.

It is suggested that the term flute be used as a genetic rather than a descriptive term, and be restricted to long parallel-sided ridges which reflect accurately the direction of ice movement and which form when deformable subglacial materials are intruded into tunnels which tend to open up on the lee sides of single rigid obstructions on the glacier bed.

In an analysis of the formation of flutes by subglacial cavities the flute-forming sediment is treated as a linear isotropic elastic material prior to failure, which occurs when Coulomb's law is satisfied. The maximum height of a flute is derived as a function of the mechanical properties of its material and the factors controlling the variation of height with distance along the flute are analysed. The flow pattern of material during the formation of a flute which the theory suggests is similar to the patterns which have been determined in a till fabric analysis of flutes near Breiðamcrkurjökull, Iceland.

Altitudes, lengths, areas, maximum thicknesses and volumes of 27 glaciers that built up in a small area of the Scottish Highlands during the Loch Lomond (Younger Dryas) stadial are given. The influence of direct radiation on the glaciers is calculated taking into account transmissivity of the atmosphere, glacier aspect and surface gradient, and the albedos of ice and snow. Equilibrium firn lines for the glaciers are determined. The computed regional firn line, incorporating the influence of direct radiation had a gradient of 14.5 m/km. Former precipitation values are based on an equation for average annual mass balances of the glaciers during the stadial that incorporates glacier altitude, regional ablation gradient, direct radiation, the influence of avalanching and blowing of snow, and final glacier volume. Average July and January sea-level temperatures of 6°C and at least — 8°C, respectively, are inferred. Snowfall was principally associated with south-east winds and was similar in total to present day but rather differently distributed. Cloud amounts, at least in summer, were greater than now. A more vigorous atmospheric circulation with many depressions moving along more southerly tracks than now may have been related to the junction of polar water and North Atlantic Drift water in the immediate vicinity of the British Isles.

A simple, relatively inexpensive system for signal recovery in retrospect was designed to record lake-ice seismic events. The system is broad-band, D.C. to 100 Hz, light-weight and may be added to any existing data-acquisition equipment. The system may be used ( 1 ) to initiate high-speed paper recordings in the field, (2) to store transient events on magnetic tape while in the field, or (3) for the unattended laboratory-recovery of transient events from a continuous field recording. Continuous, high-resolution, broad-band recordings of seismic events occurring in lake ice over a period of weeks may be stored on a single reel of magnetic tape. The initial cost of the device is rapidly equaled by the savings in tape cost and man hours expended in signal recovery.

Observations and measurements made on the Ross Ice Shelf during 1974 suggest that rime and surface hoar deposited on the surface during fogs may contribute 5-10% of the annual accumulation. Layers formed during such fogs may be mistaken for depth hoar layers in the snow-pack.

Shreve (1962) evaluated the performance of solid-nose hotpoints of various geometries. A more concise method for analysis of the special case of a flat disc melting into ice is presented here. In contrast to the numerical solution presented by Shreve, this method results in a closed analytical solution, based on constant properties, which is corrected using a property ratio scheme. This technique allows the substitution of various fluid properties into the solutions as might be required when considering contaminated or saline ice or other liquids. Shreve's method would require some additional numerical work to evaluate various properties because the temperature dependence of the viscosity of water is integrated into the results.

Results obtained for the flat disc using this method are shown to compare favorably with Shreve's, indicating that this type of analysis may be applicable to other melt-unit geometries.