Heat input to basal ice at subglacial low-pressure regions, such as exist on the lee side of bed bumps including regions of ice-bed separation, is shown to melt basal ice internally in a narrow boundary layer at most centimeters thick. Before ice at the ice-bed interface can begin to melt, the heat input Q must exceed a critical value Q*. Q* increases rapidly with an increase in the difference ΔΡ between the nominal (global) overburden pressure and the magnitude of the (local) normal stress acting between the ice and bed or ice and water pocket. Because of the non-linear nature of the flow law, the thickness of the boundary layer decreases rapidly with increasing ΔΡ. The ice in the boundary layer is likely to be soft with a high water content. Under certain conditions, a regelation cycle may exist between the boundary layer and the water in a subglacial cavity. The boundary layer is sufficiently narrow that the processes can reach steady state while ice traverses subglacial low-pressure regions of length the order of 0.01–0.1 m. The regelation phenomenon may preserve or aid the formation of narrow debris-rich ice layers at the base of temperate glaciers.

A hydrodynamic model of interface stability in a stratified fluid is reviewed. The model predicts that irregularities on the boundaries of a stiff layer, embedded in a soft matrix, are unstable in pure shearing flow, when compression is normal to the layer. Perturbations on such a layer can grow to form symmetric pinch-and-swell structures called boudins. The model predicts initial perturbation growth rates on the boundaries of an interglacial period ice layer. We find that, beneath an ice divide, irregularities on the Sangamon layer boundaries will not kinematically decay, as the layer thins. Finite-element modelling is used to determine the strain history of Sangamon ice beneath the divide at Summit, Greenland. That history suggests boundary irregularities have grown, relative to layer thickness, at least 26 fold over the past 90000 years. The result may be severe distortion or severing of the layer. Core holes penetrating the layer may recover anomalously thick or thin columns of ice resulting in erroneous environmental and climatic interpretations. Radio echo-sounding may be useful in searching for zones of boudinage, which should be avoided when coring. Initial perturbations might arise from mass-balance spatial variations or from transient flow fields.

Digital terrain models (DTM) are a very important tool for all sorts of glacier studies and serve as a basis for many applications. The latest development is the combination of DTM with digital image-processing techniques which enable a much better visualization of the DTM, and thus a better interpretation of glaciological phenomena.

These new technologies have been applied in an investigation of the Vaughan Lewis Icefall, Juneau Icefield, Alaska. A DTM has been produced by means of photogrammetrie measurements, and a glacier-flow velocity of up to 5.70 m d−1 in the steepest part has been derived. The possibilities for the interpretation of the glacier topography have been significantly improved by using digital image-processing algorithms for the visualization of the DTM.

This study investigates the processes of ice-marginal sedimentation in Vestfold Hills, Antarctica. Most debris is released from the ice when basal and englacial debris bands become warped and reach the surface of the glacier and where the debris bands are exposed by ablation of the ice surface. Once released, the debris is redistributed in the ice-marginal area by depositional processes that are controlled by the availability of water. During the short summer, melt water from snow and ice saturates the newly released debris and causes sediment flows and other mass-movement deposits. Melt-out and sublimation tills form after the layer of debris on the moraines is consolidated and melting rates decrease. When the thickness of deposits on the surface of ice-cored moraines reaches or exceeds the depth of summer thawing, the ice core no longer melts and the moraines become semi-permanent features. The sediments and land forms of the ice-marginal area closely resemble those formed by sub-polar glaciers with a complex thermal regime and are unlike those that form at the margins of dry-based polar glaciers. Although glacier thermal regime is understood to be a major control on debris dispersal and processes of glacial sedimentation, the evidence from Vestfold Hills suggests that the primary control is the climate of the glacier terminus area.

Drilling bore holes in deep, cold ice masses by hot-water methods and maintaining these holes with sufficient diameter to allow down-hole experimentation poses a major obstacle to the investigation of conditions beneath ice sheets and ice streams. Closure of the water-filled holes by refreezing is the dominant difficulty. In this paper, we describe calculations of heat transfer from the drilling system to the ice and the subsequent time-dependent motion of the phase boundary defining the bore-hole wall. Results are presented with the view of optimizing the bore-hole radius at depth for a fixed drill performance and a variable rate of drilling.

Calculation of melting/refreezing rates at the bore-hole wall requires the use of a one-dimensional, time-dependent numerical heat-flow model with a distorting mesh which follows the changing hole size. The delay of hole closure is discussed with a view to keeping holes open long enough to allow instruments to be lowered to the glacier bed, while realizing that drilling-system performance may be marginal because of logisitical and/or expenditure constraints. The relative merits of drilling a large hole, which is very time consuming with a small drill, and the use of water-soluble antifreezes, which have a history of creating plugs of ice slush, are discussed. A method of creating a stable hole filled with antifreeze in which ice slush does not occur is described.

The recent application of these theoretical ideas to the planning and implementation of successful hot-water drilling programs in Antarctica and Greenland is also presented.

Air content (V) of polar ice has been used as an indicator of the past elevation of the ice sheets. A calculation is presented to correct V measurements performed on ice samples for the effect of cut bubbles at their surface. The results indicate a correction ranging from 1 to 10% for cubic ice samples with about 3 cm length. The correction depends mainly on the size of the bubbles. The theoretical calculation is experimentally verified. The statistical noise linked with the presence of a finite number of bubbles in the ice samples is evaluated. The influence of such a correction on the V profiles measured on polar ice cores is discussed. The method in this paper can also be used for correction of ice-density data obtained by the hydrostatic method.

Glacial abrasion was simulated in experiments in which a small artificial glacier bed was pushed beneath a fixed ice block under pressure. The experiments provide a means of testing theoretical models of abrasion, particularly those factors that govern the magnitude of stress concentrations beneath abrading rock fragments. In preliminary experiments, vertical ice flow around a sphere mounted on the bed was studied. In subsequent experiments, marble tablets were pushed beneath granitic rock fragments frozen into the base of the ice block. Unlike previous abrasion experiments, the sliding velocity was realistic (25 mm d−1), and ice near the bed was at the pressure-melting temperature. Resultant striations closely resemble those observed on glaciated bedrock.

As predicted by Hallet (1979), the component of the ice velocity towards the bed strongly influenced stresses beneath fragments, and classical regelation and creep theory provided an approximate estimate of the downward drag force on fragments. Half of the rock fragments rotated significantly, accounting for 10–50% of their motion relative to the bed and influencing abrasion rates and the shear stress supported along the ice-bed interface. Striation patterns indirectly suggest that fragment rotations were inhibited by increases in ice pressure, which presumably increased the drag on roughness elements on fragment surfaces. This may have resulted from a reduction in the thickness of the water film around fragments, facilitated by leakage of water from the bed.

We present data on ice texture, salinity, and δ18O obtained from identical sections of ice cores during the Winter Weddell Sea Project 1986 on RV Polarstern from July through August 1986, in the longitude range between 5°W. and 7°E. We find no uniquely definable relationship between δ18O values and ice texture in a particular section. However, most of the snow ice as well as some sections of frazil ice are found to have negative δ18O concentrations. This is due to varying degrees of admixtures of meteoric ice (snow) and sea-water during formation of snow ice. In contrast to common assumptions, our results seem to indicate that a snow cover contributes positively to sea-ice growth rather than slowing down the overall growth rate. Based on a simple model, we have estimated the contributions of meteoric ice (mean of 3 ± 3%) and the combined meteoric ice/sea-water fraction (a minimum of 7 ± 6%) to the total ice thickness for the majority of the sampled floes. Although this is only a moderate contribution to the overall mass balance, in the absence of congelation growth it nevertheless enhances ice growth in general. This hypothesis is independently supported by our snow- and ice-thickness data (Wadhams and others, 1987), which demonstrate that the depression of the snow/ice interface below the water line (i.e. a negative freeboard) and the formation of snow ice is a common occurrence in the Weddell Sea. Therefore, we hypothesize that the major part of the observed apparent increase in ice thickness between our inbound and outbound tracks of WWSP’86 may not be derived from “regular”, thermodynamically driven congelation growth, but rather from the snow-ice component in floes of the Weddell Sea.

Ice flow along the 20 km long strain network up-stream of the Dye 3 bore hole in Greenland is studied in detail. By solving the force-balance equations and using selected flow laws, stresses and strain-rates are calculated throughout the section of the ice sheet. The validity of the results is evaluated by comparison with the velocity profile derived from bore-hole-tilting measurements, and with observed surface strain-rates. A number of constitutive relations are tried and most predict a velocity profile at the bore-hole site that is in good agreement with that observed, if appropriate enhancement factors are used. However, there are major discrepancies between modelled and measured surface strain-rates. Use of Nye’s generalization of Glen’s flow law, or an anisotropic constitutive relation, requires unrealistically large along-flow variations in the enhancement factor. Inclusion of normal stress effects can lead to much better agreement, but it is possible that other processes, such as dynamic recrystallization or primary creep, should be included in the constitutive relation of polar ice.

Results of pollen analysis of a large firn sample from Dye 3, southern Greenland, are reported. The sample was divided into three sections: all dominated by pine, birch, and ragweed pollen. Pollen concentration for the entire sample is 15 pollen grains per litre and the annual pollen deposition rate has been estimated at approximately 0.73 grains cm−2 year−1 (7300 grains m−2 a−1). Although these values are slightly lower than expected, considering the relatively southern latitude of the site, they are within the range of values obtained from other Arctic ice caps.

Coefficients of thermal linear expansion were determined for sea ice using a Michelson interferometer. Over a temperature range of −4 °to −15 °C, the coefficients varied from 45 ×10−6 to 54×10−6 °C−1 for ice with a salinity of 2 ppt, and from 33 ×10−6 to 53 ×10−6 °C−1 for ice with a salinity of 4 ppt. Initially, warming the sea ice resulted in coefficients that were the same as those for fresh-water ice, within the limits of experimental error. Subsequent sea-ice cooling resulted in coefficients that were initially lower than those for fresh-water ice, but that asymptotically approached the coefficient values for fresh-water ice at colder temperatures. On the second warming and cooling cycle, the coefficients of thermal linear expansion exhibited hysteresis and a decrease in magnitudes. We have also shown that Pettersson’s (1883) and Malmgren’s (1927) measurements of the thermal volume expansion of sea ice were the result of phase transitions that caused brine expulsion, when air-free sea ice was cooled, and internal porosity increases, when sea ice was warmed.

Our results indicate that Petterson’s (1883) and Malmgren’s (1927) measurements of the thermal volume expansion of sea ice are in error. Consequently, theoretical descriptions based on their results are incorrect (Anderson, 1960; Zubov and Savelyev (given in Doronin and Kheisin (1977)); Doronin and Kheisin, 1977). Our results for the initial sea-ice warming cycle do agree with Cox’s (1983) analysis.