Three satellite-tracked data buoys were deployed between 70° and 80°E south of 65°S in February-March 1985. These buoys were subsequently trapped within the expanding seasonal sea ice and drifted with the ice. The buoys measured air temperature and pressure, and water temperatures to 100 m depth. Data from the buoys are used to describe the ice drift and environment within the winter sea-ice zone in this region, both north and south of the Antarctic Divergence. Additional preliminary data from a further six buoys deployed in the same area in March 1987 are also presented.

Besides providing information on the broad-scale drift of the ice in the Prydz Bay region, data from the buoys have shown: (i) the important role that ice drift plays in determining the autumn and winter expansion of Antarctic sea ice; (ii) the highly mobile nature of the ice, even hundreds of kilometres from the ice edge; (iii) the role that ocean-bottom topography has in determining ice drift over the continental shelf; and (iv) the modifying influence that an ice cover has on regional climate.

A qualitative assessment is made of the relative importance of the major forces driving the ice, although the data are insufficient for a detailed study of the ice dynamics.

New data on leads and surface-melt phenomena in the Arctic, based on mapping of DMSP visible images, are presented. Lead orientations in the Beaufort Sea are broadly correlated with geostrophic wind direction and show similar synoptic scale patterns. Preliminary results of airborne 1.06 μm lidar transects over Baffin Bay demonstrate its great potential for high-resolution mapping of open-water areas and, in winter/spring, their ice-crystal plumes. Some of these sub-visible plumes are observed to penetrate the Arctic inversion. Snow-melt maps for the entire Arctic prepared for four summer seasons have been used to derive surface-albedo data. From these data the variability of the surface-energy balance is estimated to be equivalent to 0.6 m of ice melt.

A three-dimensional dynamic, thermodynamic ice-sheet model has been developed to simulate the past, present, and future behaviour of the Antarctic ice sheet. The present ice velocities depend on the deep ice temperatures which in turn depend on the past changes of the ice sheet, including surface temperature, accumulation rate, and ice thickness. The basal temperatures are also strongly dependent on the geothermal heat flux. The model has therefore been used to study the effect on the basal temperatures, of changes to the geothermal heat flux, as well as the past changes of surface temperature and accumulation rate based on results obtained from the Vostok deep ice core. The model is also used to compute the distribution of surface velocity required to balance the present accumulation rate and the dynamics velocity based on the stress, temperature, and flow properties of ice, for the internal deformation, plus a component due to ice sliding. These velocities are compared to observed surface velocities in East Antarctica to assess the state of balance and the performance of the dynamics formulation.

Single folds which enclose the marginal “inner moraines” of the dry-based Jeremy Sykes and Alberich Glaciers, Dry Valleys area, Antarctica, are deduced to have formed during an advance of the margins of these glaciers where the thin, stiff marginal ice has obstructed glacier flow and expansion.

It is suggested that the occurrence of this type of fold can be an indicator of an advance in many fully dry-based glaciers. On Alberich Glacier, the fold and associated margin forms suggest that a sequence of margin forms evolves when a small cirque glacier expands outward from a sloping ramp front to a tongue terminating in a calving ice cliff.

The ice divide on an ice sheet is a special place on the ice sheet, where the velocity and stress distributions are different from those found just two ice thicknesses from the ice divide. A two-dimensional thermo-mechanical model has been used to model the flow near the ice divide in central Greenland. The results show that the surface strain-rates are increased by 50% of the values found 15 km down-stream and that the basal temperatures rise by 3°C. Estimates of a depth–age distribution in the vicinity of Crête show that the isochrones rise at the dome. The model calculations predict 1600 m of Holocene ice, 1000 m of Wisconsin ice, and 400 m of ice older than 115kyear. 15 km down-stream from the dome, the thickness of ice older than 115 kyear will be reduced by 50%. In these calculations it has been assumed that the ice thickness at the dome is time-invariant. The existence of the old ice depends on the basal temperature. The estimates of the basal temperature depend strongly on past accumulation rates and on the geothermal heat flux but the ice has most likely been below the pressure-melting point throughout the last glacial period.

Glaciological measurements from Austfonna on Nordaustlandet, Svalbard, are needed as a prerequisite to mathematical modelling of ice-mass dynamics. Several upper and lower boundary conditions are set out in detail for a 670 km2 drainage basin (Basin 5) and are generalized to the whole ice cap where possible. The ice surface and bed topography are mapped for Basin 5. 30% of the basin lies below sea-level. Bed elevations range from -100 m to over 300 m, and maximum ice thickness is >500 m. A 21 km long trilateral network of stakes provides velocity and strain-rate data. Maximum ice-surface velocity is 47 m a−1 and maximum strain-rate is 0.64 × 10−2 a−1. Snow-line migration with time is mapped from digital Landsat MSS data, and mass-balance estimates are used to calculate balance velocities. At the equilibrium line, about 300–350 m in elevation, balance velocity and observed ice-surface velocity are comparable, indicating that the basin is approximately in balance. A first approximation is given for the rate of iceberg calving from the tide-water basin margins. Enhanced Landsat imagery also shows that turbid melt-water plumes of subglacial origin flow from the terminal ice cliffs, indicating that at least parts of the ice-cap margin are at the melting point. The margins of Basin 5, grounded below present sea-level, are likely to be underlain by deformable sediments, but inland the nature of the substrate is unknown.

The internal dynamics of the Law Dome ice cap have been investigated by measuring the deformation of three bore holes located on an approximate flow line. Bore holes BHC1 (300 m deep) and BHC2 (344 m) were drilled in the coastal area to within several metres of bedrock and BHQ (418 m) was drilled about half-way towards the dome centre to about 50% of the ice thickness. Detailed measurements of orientation (inclination and azimuth), diameter, and temperature were taken through each bore hole over a 1 year span for BHC1 and BHC2 and a 10 year span for BHQ. The orientation data were reduced to obtain ∂u/∂z, a measure of the shear strain-rate. Changes in the depth of features located by bore-hole diameter measurements were used to obtain vertical velocity profiles. Other measurements discussed are temperatures, oxygen isotopes, crystal structure, surface velocities, and surface and bedrock topography.

At the coastal sites, the ∂u/∂z profiles show two maxima in the lower third of the ice sheet. Flow due to the measured deformation accounts for about 55% of the surface velocity, the remainder being due to deformation and sliding in the basal zone. The vertical velocity profiles show mostly firn compression. The deeper ∂u/∂z maximum occurs in ice from the Wisconsin period which appears to deform more rapidly than the Holocene ice immediately above. The upper ∂u/∂z maximum may be related to the stress history of the ice, which can also explain the presence of significant shear strain and crystal-fabric development at only half the ice thickness at the BHQ site.

Results from movement surveys on Rutford Ice Stream are presented with complementary surface-elevation and ice-thickness measurements. Surface velocities of 300 m a−1 occur at least 130 km up-stream of the grounding line and contrast strongly with the neighbouring Carlson Inlet, where a velocity of 7 m a−1 has been measured. This contrast in velocity is not topographically controlled but appears to be due instead to differences in basal conditions, with Carlson Inlet probably being frozen to its bed. Concentration of lateral shear close to the margins and surface expression of subglacial topography both support a view of significant basal shear stresses in the central part of Rutford Ice Stream. The pattern of principal strain-rate trajectories shows a small number of characteristic features which can be compared with results from future modelling of the glacier's flow.

A dimensionless model of thermo-mechanically coupled ice sheets is used to analyse the operation of the system. Three thermal processes are identified: (i) dissipation, having a maximum time-scale of thousands of years; (ii) advection, having a time-scale of tens of thousands of years; and (iii) conduction, having a time-scale of 100000 years. Kinematical processes occur on two time-scales: (i) a marginal advective time-scale of thousands of years; and (ii) a diffusive time-scale of tens of thousands of years dominant in the divide area.

The coupling with the temperature field in the bed produces fluctuations to the depth of a few kilometres, which means that horizontal conduction in the bed can be ignored except perhaps in the marginal area. The thermal inertia of the bed could produce significant fluctuations in the geothermal heat gradient.

The operation of the thermo-mechanically coupled system is explored with a time-dependent thermo-mechanically coupled numerical algorithm. Dependence of the basal friction on temperature is introduced heuristically, and an enthalpy method is used to represent the effect of latent heat. The marginal area is shown to be dissipation-driven, and always reaches melting point. The divide area can show two modes of behaviour: a warm-based mode where the ice sheet is thin, and a cold-based mode where the ice sheet is thick. Which mode operates depends upon the applied temperature field and the amount of heat conducted from the bed.

Calculations where sliding is limited were not found to be possible owing to problems with the reduced model which resulted in a violation of the approximation conditions at the margin. Cases which did work required a substantial sliding component; as a result, a significant coupling between geometry and temperature can only be demonstrated when sliding is made temperature-dependent.

The zero isopleth of surface areal strain-rate is roughly parallel to, but below, the equilibrium line of the glacier. In the middle and lower parts of the ablation area, the basal temperature is either close to or at the melting point, whereas above and around the equilibrium line the glacier is frozen to its bed. The basal bulk ice is deforming continuously and inhomogeneously but, in the inner parts of the glacier, where the strain-rate is smaller, homogeneous deformation can only be assumed conditionally. In the ablation area, pure shear is predominant near the base of the glacier and simple shear becomes progressively more important from the inner parts of the glacier towards its margins.

Calving ice walls are an important ablation mechanism for deglaciation of calving bays occupied by temperate tide-water glaciers and polar marine ice sheets. Dangers inherent in calving bays have precluded detailed field studies of these calving ice walls. However, calving ice walls also exist in sub-polar glaciers terminating on dry land, and an opportunity for detailed field work was afforded by the 12 August 1970 volcanic eruption on Deception Island (63.0°S, 60.6°W), where thawing of a surface blanket of ice-cemented ash produced solifluction ramps that made parts of the ice wall accessible. Measurements made in a melt-water trough incised into the ice wall, and in four tunnels cut into the ice wall, revealed numerous shear bands that rose almost vertically and curved forward. Shear offset increased upward and was greatest in shear bands that intersected the tips of ring-fault crevasses on the up-slope side of the ice wall. Near the base of the ice wall, other shear bands, possibly related to the slip-line field, intersected the ice wall at about 45°. Ice slabs separated by ring faults calved straight down as a result of shear rupture along these two sets of shear bands. Calving dynamics were analyzed and generalized for ice walls grounded in water.

A flow relation for the minimum strain-rates for isotropic ice is presented. This flow relation has been compiled from laboratory tests in shear and compression, and covers the temperature and stress ranges typical of natural ice masses. However, the ice in natural ice masses is mostly anistropic. Three cores drilled at Law Dome, Antarctica, are used as examples to illustrate some of the different anisotropies that develop in ice masses in relation to the stress configurations and accumulated strains.

For the shear- or compression-stress configurations, anistropic flow rates can be derived as enhancements of the minimum isotropic flow rates for the particular stress, temperature, and stress configuration. For compression, this enhancement asymptotically approaches 3. For shear, the enhancement may be as high as 8. For combinations of shear and compression, further tests are required.

The shape, surface features, composition, and thickness of icebergs trapped annually in a 200 km long coastal strip of fast ice have been examined to determine their sources and movement. The thin western ice front of the Ross Ice Shelf seems to produce about 40% of the icebergs while local glaciers produce the remainder. The ice-shelf icebergs are carried west towards Ross Island then north up the western side of the Ross Sea. A small proportion of them gets trapped mainly by grounding on shallow areas of the sea floor which protrude across the regional long-shore currents.

The surface movement of part of a Nanga Parbat (Himalaya) glacier immediately below an ice fall has been investigated by terrestrial stereophotogrammetry. This method provides an opportunity of measuring many more points than by using other procedures. The shape of the transverse velocity profile indicated normal stream-like flow and not “blockschollen movement”, as might have been expected, because of its high velocity (45 cm/d). But, one feature of the ice fall above had been carried down-glacier to the location of the profile; namely, several discontinuities along the velocity diagram. The ice fall had transmitted a small part of its splitting into seracs to the flatter area below, but not its blockschollen movement.

There seems to be a contradiction in that on one side there are discontinuities, i.e. the glacier is torn up into blocks by shear planes, but on the other side the glacier moves en bloc. In fact, a short-term investigation over hours or just a few days discloses cracks but, after time periods of a week or more, these irregularities compensate for one another so that they merge into a single unit block motion.

We report on the development and physical properties of sea ice in the central and eastern Weddell Sea. The investigations were part of the Winter Weddell Sea Project 1986, which extended over the months of July through December. Major elements of the glaciological part of this study included continuous shipborne observations of sea-ice conditions and occasional helicopter reconnaissance flights, extensive measurements of snow and ice thicknesses at daily ice stations, and detailed analyses of sampled ice cores from each ice station. Textural investigations of the sampled ice revealed the dominance of frazil ice in the central Weddell Sea and the occurrence of an additional ice class, called platelet ice, together with the commonly known frazil and congelation ice in the coastal region of the eastern Weddell Sea. These results, in combination with the visual ice observations, reveal two major mechanisms for sea-ice generation in the Antarctic, which were not sufficiently well accounted for in previous investigations. In the central Weddell Sea, a cycle of pancake-ice formation and its growth into consolidated floes seems to be the dominant process of the advancing sea-ice edge. In the coastal waters, the growing sea-ice cover consists, to a considerable degree, of ice platelets which are formed in the underlying water column in front of the ice-shelf edges. Thus, congelation-ice growth, which is mainly controlled by atmospheric, thermodynamic forcing, seems to be of less importance in the central and south-eastern Weddell Sea than, for example, in the Arctic Basin.

We present model calculations that describe the flow of ice shelves of different configurations. We consider “ideal” models with well-defined boundary conditions and simple geometry in order to explore the response of an ice shelf to transient variations in ice-stream input. Gradually increasing the complexity of these simple models allows a better understanding of ice-shelf behavior without the complications that would arise in considering natural ice shelves. We find that the dissipation of ice-thickness variations caused by ice-stream transience is strongly influenced by ice rheology. The presence of an ice rise significantly alters the velocity field of the adjacent ice, when changes in ice-stream input occur. With models of “real” ice shelves, we demonstrate the ability of numerical models to test successfully working hypotheses on ice-shelf thickness distributions. Ice velocities, obtained by diagnostic models of Filchner–Ronne Ice Shelf that use different ice-thickness distributions, are compared with measured ice velocities. This comparison demonstrates that the model employing regions of thin ice in the central part of the ice shelf yields velocities significantly different from the field data. We therefore conclude that zones of thin ice on Filchner–Ronne ice Shelf are unlikely. This conclusion has recently been confirmed by field measurements.

As part of the Winter Weddell Sea Project 1986 (WWSP 86), a buoy, transmitting via TIROS-N satellites to Service Argos, was inserted into an ice floe in the southern Weddell Sea. Operational U.K. Meteorological Office numerical surface-pressure analyses, which utilized the buoy’s measured values of air pressure and temperature, are used to assess the impact of weather systems on pack-ice movement. The motion of the buoy is shown to be related closely to the position of the circumpolar trough and to the tracks of depressions crossing the area. The tracks of this and other buoys deployed during WWSP 86 are analysed, together with the known drifts of some ice-bound vessels, to establish the overall movement of sea ice in the central and western Weddell Sea. Using these data, the area of ice transported northward out of the Weddell Sea is determined. Roughly 60% of the winter sea-ice cover is discharged out of the area, and is replaced by new ice formation in coastal polynyas and by influx of new ice from the east. In summer, a further 30% is discharged northward out of the region, leaving 40% cover and by implication a 30% loss by melting.

Passive microwave data (SMMR) and data from drifting buoys are used to examine atmosphere-ice interactions during autumn 1978 through autumn 1982 in the Canada Basin. SMMR brightness temperatures, ratios, and calculated ice concentrations are compared with surface pressure, temperature, buoy motion, and winds at a coincident location to identify possible linkages operating on daily to weekly time-scales. Inter-annual differences in temperature, pressure, and ice conditions inferred from SMMR data are discussed. Periods of reduced ice concentration as estimated by SMMR are found to occur during late summer and early autumn in 1980, 1981, and 1982. The decreases in ice concentration are concurrent with low-pressure systems and increased divergence in the buoy field in 1980 and 1981. These apparent decreases in ice compactness are discussed in relation to previously reported areas of reduced concentration interpreted from ESMR data. Evidence to support a postulated relationship between atmospheric low pressure and divergence of ice under free-drift conditions is demonstrated for the summer of 1980.

Detailed analysis of radar images, obtained using a synthetic aperture radar technique, from the bottom surface of a floating ice sheet suggests the presence of linear, groove-type scarring. The dominant physical parameters, such as the orientation, depth, and periodicity of the underside scarring, are derived. Although these results relate to the bottom roughness of Bach Ice Shelf (Alexander Island), Antarctica, close to its line of grounding, they are comparable to those of the underside of Ross Ice Shelf. It therefore seems probable that the roughness parameters, measured on Ross Ice Shelf using standard radio-echo techniques, relate to linear grooves which were introduced in the region of ice grounding. Such scarring runs along the direction of ice flow and is thought to be a drawn-out imprint of the bedrock surface where the ice was last grounded.

A large part of the area of the Shirase Glacier drainage basin has been surveyed by airborne (operating frequency: 179 MHz) and ground-based (60 MHz) radio echo-sounding to define the bedrock topography and to investigate the condition of bed/ice interface since 1982.

It is shown that the reflection intensity from the bed, which is corrected for attenuation in the ice sheet, has a higher value for reflection intensity in the down-stream area of Shirase Glacier than in the up-stream area. The area of strongest intensity of reflection from the bed coincides with the area for which the calculated temperature at the bed is above −1°C. The boundary area between the highest and lowest values of corrected reflected intensity corresponds to the area of decreasing basal shear stress. It is found that the distribution of high corrected reflection intensity corresponds to the area of thinning of the ice sheet, which has been measured by ice-flow observation in the Shirase Glacier drainage basin.