Pit and core studies at the Upstream Β camp (UpB) and ridge BC (BC) on the Siple Coast, West Antarctica, have revealed a number of interesting results. Both sites have 10 m temperatures near −26.5°C and accumulation rates near 0.09 m a−1 of ice. At both sites, the low densities of annual depth-hoar layers arise in the first year following snow deposition. Densification rates at UpB are accelerated compared to BC and to other sites, probably owing to stress enhancement of power-law-creep densification caused by the large longitudinal deviatoric stresses at UpB. The connectivity of firn, measured by the number of bonds per grain and by the fraction of total surface area per grain involved in bonds, correlates well with density and shows no significant dependence on grain-size. The longitudinal stresses at UpB cause the ice to develop a fabric of horizontally elongated bubbles and grains, interpenetrating grains, and strain shadows between about 30–70 m. Ice at BC below about 95 m shows rapid grain growth and development of a bimodal grain-size distribution, but no preferred orientation of grains or pores. The temperature-depth profile to 100 m at BC suggests a basal heat flux near 1.9 heat-flow units (HFU), typical of a region of active rifting. Surface-melt events occur about once every 50 years, and correlate well between UpB and BC.

Surface velocity and deformation, radar sounding, and aerial photography data are used to describe the flow of Ross Ice Shelf around Crary Ice Rise. A continuous band of crevasses around the ice rise now allows the complete boundary to be mapped for the first time. The dynamics of three distinctly different areas of ice flow are studied. Just up-stream of the ice rise, there is a region of ice rumples dominated by intense longitudinal compression (0.01 a−1) and lateral tension. On the south-west side of the ice rise, intense shear (0.03 a−1) dominates, with the boundary layer of affected ice-shelf motion extending over 20 km from the ice-rise edge into the ice shelf. North-west of the ice rise, a crevasse-free block of ice, 40 km × 7 km, appears to have separated from the main ice rise and is now moving with the ice shelf. We refer to such moving blocks of ice as rafts. The separation of this raft is calculated to have occurred 20 ± 10 years ago. Other possible rafts are identified, including one on the south-west side of the ice rise which appears to be in the process of separating. Mechanisms for the formation of rafts are discussed.

In January 1986, reflection-seismic measurements were made in the central part of Filchner-Ronne Ice Shelf, north of Henry and Korff ice rises, by a German oversnow traverse. With the help of a newly developed “ice streamer”, two people were able to obtain seismic records nearly every 5km over a distance of 150km within 3d. Processing of digitally recorded seismograms yielded profiles of ice thickness and sea depth. The ice thickness varies from 400 to 500 m and is in good agreement with the thickness of 465 m at a drill hole. The reflection-seismic results are comparable with those obtained from aerogeophysical measurements.

Survey data from three transverse profiles across Rutford Ice Stream show that most of the shear deformation occurs in boundary layers about 10 km wide on both sides of the glacier. There is a central part, also about 10 km wide, where lateral shear-stress gradients are not significantly different from zero. This implies that friction from the side walls does not impede the flow in the middle of the glacier. Instead, a basal shear stress of around 40 kPa is required in order to provide the necessary restraint. Even then, only a small proportion of the forward movement will be caused by internal deformation of the ice and the remainder must arise from sliding or rapid deformation at or near the base of the glacier. Although the present analysis does not consider in detail the variation with depth in stresses and strain-rates, the formulation of the problem suggests that a self-consistent solution can be found for the distribution of basal and vertical shear stresses.

Accumulation rates in the eastern part of Ronne Ice Shelf were determined by isotopic stratigraphy (18O). The samples were taken from snow-pits dug during the Filchner I and II operations in 1984 and 1986. In general, the accumulation rate decreases towards the south; the greatest decrease, from 21.3 to 13.3 g cm−2 a−1, was observed between Filchner Station and measuring point 341, sited 270 km up-stream of the ice edge. The δ18O values of the near-surface layers vary between −25 and -29‰. The 18O content in the more southerly part is progressively depleted in the direction of Möllereisstrom, paralleling a decrease in the accumulation rate. Near the ice edge the 18O content decreases to the west. A 100 m ice core drilled in 1984 at point 340, 220 km from the ice edge, probably goes back to A.D. 1460; it has been dated by isotopic stratigraphy.

The accumulation rate up-stream of the drilling site was deduced from the sequence of annual layers, using a simple ice-flow model. The accumulation shows strong variations over the last 200 years, which may be caused in part by local variations in the accumulation on Ronne Ice Shelf.

A preliminary version of a three-dimensional ice-sheet model for later use in climate models, but excluding ice shelves and basal sliding, is presented and applied to the Antarctic ice sheet. In the model, the three-dimensional fields of velocity and temperature are calculated in the coupled mode, and the temperature equation is integrated for 150 000 years; the shape of the Antarctic ice sheet remains fixed. The results from the model are consistent with a stationary state in the central parts of the Antarctic ice sheet, but not in marginal areas, where the flow in the model is too small. Including a parameterized form of basal sliding that is dependent on the water pressure is likely to improve the situation.

The satellite positioning systems, NNSS or Transit system and NAVSTAR-GPS, are used successfully for the determination of ice motion. The ice motion is derived from the change in the coordinates of a station between at least two measurement epochs. Simultaneous satellite observations on solid-ground-based and on ice stations yield precise relative or local coordinates between the stations. The ice motion can be determined very accurately from the variation in these coordinates. The field observations and post-processing steps, which differ slightly for the two positioning systems, are outlined. Results for several examples are presented. Ice-motion solutions are discussed for data from Anvers Island (NNSS), from Filchner-Ronne Ice Shelf (NNSS, GPS), and from Ekström Ice Shelf (NNSS, GPS). Slow velocities of a few dm/d could be estimated in the course of one field season and the velocity values found are confirmed by annual station displacements.

In 1985–86, western Neuschwabenland was surveyed with an electromagnetic reflection system (EMR) over flight tracks totalling 6100 km. The area spans a triangle from Ekström Ice Shelf and Heimefront Range to the Kraul Mountains [Vestfjella] and covers 48 000 km2.

Data on surface elevation, ice thickness and bedrock topography are presented in maps. The bedrock-elevation map shows more detail than earlier compilations, with a remarkable graben-like structure north of the Heimefront Range and large areas of bedrock below sea-level.

In the same season, tracks totalling 1600 km were flown over Berkner Island. High electromagnetic reflectivity was found over the southern part of Berkner Island, whereas the northern part showed considerably lower reflectivity. We believe that this is due to melting conditions in the southern part and freezing conditions in the northern part.

Ice sheets consist of several disjoint regions, each with physically distinct behavior. Parts are cold; others are temperate and either partly or completely saturated. Varying dust content, impurities, debris content, etc., may affect the ice flow. Usually, these regions are separated by material or non-material surfaces or boundaries. We use mixture concepts, involving the balances of mass, momentum, energy and entropy. When applied to regular domain points these concepts and appropriate constitutive postulates yield the field equations for the evolution of the constituents. When formulated in terms of singular surfaces, boundary and transition conditions emerge. Our presentation takes the form of an extended abstract of work that is presently under consideration (Hutter and Engelhardt 1988; Hutter and Engelhardt, in preparation).

An efficient numerical ice-sheet model, including time dependence and full thermo-mechanical coupling, has been developed in order to investigate the thermal regime and overall configuration of a polar ice sheet with respect to changing environmental conditions.

From basic sensitivity experiments, in which a schematic East Antarctic ice sheet is forced with a typical glacial–interglacial climatic shift, it is found that: (i) the mutual interaction of temperature and deformation has a stabilizing effect on its steady-state configuration; (ii) in the transient mode, this climatic transition initially leads to increased ice thickness due to enhanced accumulation, after which this trend is reversed due to a warmer base. Time-scales for this reversal are of the order of 103 years in marginal zones and of 104 years in interior regions; (iii) horizontal heat advection plays a major role in damping possible runaway behaviour due to the dissipation – strain-rate feed-back, suggesting that creep instability is a rather unlikely candidate to initiate surging of the East Antarctic ice sheet.

The model is then applied to four East Antarctic flow lines. Only the flow line passing through Wilkes Land appears to be vulnerable to widespread basal melting, due to enhanced basal warming following climatic warming. Time-dependent modelling of the Vostok flow line indicates that the Vostok Station area has risen about 95 m since the beginning of the present interglacial due to thermo-mechanical effects, which is of particular interest in interpreting the palaeoclimatic signal of the ice core obtained there.

A method for calculating the impulse response of a layered dielectric is used to investigate the effect of near-surface density stratification on the interpretation of radar-altimetry data collected over ice sheets. Results are computed for a model consisting of a measured density profile that is artificially interrupted by an ice layer inserted at successively greater depths. The incident radar signal is modeled by a 20 ns long pulse modulated at 1 GHz. Analysis using a simple scheme for estimating the one-way travel time of the wave indicates as much as a 2 ns (equivalent to 60 cm in range) apparent time delay for a density profile that has been modified by inclusion of an ice layer relative to the ice-layer-free case. This suggests that layering may be a factor in interpreting repeated altimeter measurements of the ice sheets in terms of net growth or decay.

Radar ice-thickness surveys and bore-hole measurements suggest that the central part of Ronne Ice Shelf possesses a lobe-shaped basal layer of undetermined nature (probably saline ice). This layer is characterized by high radio-wave absorbtivity and by thicknesses up to approximately 300 m. We reconstruct this basal layer and the associated ice-shelf thickness and flow distributions, using a time-dependent ice-shelf model forced with prescribed basal freezing rates. Characteristics of the basal layer are controlled by two factors: (i) long ice-column residence times in the unventilated pocket between Henry and Korff ice rises and Doake Ice Rumples, and (ii) basal freezing rates in this pocket that exceed the snow-accumulation rate (currently averaging 0.35 m/a ice equivalent across the ice shelf).

A numerical ice-shelf model is employed to observe the inception of the West Antarctic ice sheet (WAIS) from a thin (20 m thick) floating ice cover under the following conditions: (i) a lower sea-level than at present, due to ice-sheet formation in the Northern Hemisphere, (ii) surface and basal temperature and accumulation rates approximately equal to those of present Antarctic ice shelves, and (iii) ice flow from East Antarctica into West Antarctica is neglected. The model determines the flow and thickness of floating ice and assumes that grounded ice is stagnant. Under these constraints, all regions except the Ross Sea, the Filchner region (east of Berkner Island), and up-stream of Thwaites Glacier ground within 4000 years. Ice readily grounds in the Ronne region (west of Berkner Island), forcing ice from Ellsworth Land to flow east toward the Filchner region. It is suggested that grounding over the Ross Sea, the Filchner region, and up-stream of Thwaites Glacier occurs only after grounded-ice flow is established. Grounded-ice flow is also a prerequisite of bed erosion and sediment deposition, which leave historical records of the actual ice-sheet formation. It is suggested that erosion and sediment deposition is minimal over the Ronne region and considerable along the path from Ellsworth Land to the Filchner region, because more ice flows toward the Filchner region than the Ronne region. It is probably difficult for ice to ground over the Ross region, so this region should have a high proportion of glacial marine sediments.

Configurations of relict surface-crevasse bands and medial moraines that emanate from the shear margins of ice streams are simulated, using a numerical model of an ideal rectangular ice shelf to determine their potential for recording a past ice-stream discharge chronology.

Large-scale motions and strain-rates over great distances on polar ice sheets are often obtained from the tracking of Transit (or doppler) satellites. The results of different processing techniques for these tracking data are compared, using some of the data collected on and near Ice Stream C. Reduction is made by using the software packages CALIPER, GEODOP V, MAGNET, and the micro-processor on the Magnavox MX 1502 satellite receiver. The orbital data broadcast by the satellites are used, as well as more precise orbits obtained afterward. In addition, calculations are made for single sites individually (point positioning) and for many sites with simultaneous tracking data (network adjustment).

The results agree within the range of known errors associated with the orbits. Earth-based positions (latitude, longitude, ellipsoidal height), based on the broadcast orbits, agree to within 41.1 m. Positions with more precise orbits are within 0.7 m of one another. Relative positions are best obtained by using network techniques, and these agree with terrestrial survey results within 0.2 m in horizontal separation for sites 19 km apart, and are within 4.8 m in elevation difference. The calculated azimuth differs by 1.5 m/19 km or 10−4 rad.

The glacial geodetic contribution to the mass balance and dynamics of ice shelves includes repeated determinations of the absolute position (ϕ,λ,Η) of selected points (using satellite methods), the establishment of relative positions (y,x,Δh) in deformation figures, and height measurements. The results are used to establish ice-flow velocities and directions, strain and rotation rates, and changes in height. Modelling of deformation parameters at a few points over a large ice shelf is made possible by the collocation method. Results of these observations and analysis of Ekström Ice Shelf for the period 1979–87 are reported.

A 130 m long ice core has been dielectrically profiled. From an analysis of the measurements, we obtain a profile of the high-frequency (radio-echo) conductivity. This profile has been represented by a simplified 700 layer model. The model has layers of differing conductivities, permittivities and thicknesses. A reflection-coefficient log can be calculated, assuming that permittivity is a smooth function of depth. Variations in conductivity are shown to be more likely sources of internal reflections from depths greater than a few hundred metres than changes in permittivity caused by density changes. The log is convoluted with input pulses of various frequencies and pulse lengths in order to produce synthetic radargrams. These show features that correspond to the internal reflections typically seen when radio echo-sounding polar ice sheets.

Measurements of the bore-hole diameter were made with three-contact-point calipers from the shallow layer to a depth of 700 m at Mizuho Station, East Antarctica, in 1984–86. The minimum strain-rate (i.e. secondary creep rate) of the bore-hole closure was estimated from the observed creep curve at each depth. A relation between stress and strain-rate was deduced for a temperature of −35°C and a stress range from 0.8 to 1.65 MN m−2, as

where is the effective strain-rate (s−1) and τ is the effective shear stress (MN m−2). This result shows that, for comparable overburden stresses and temperature, the bore-hole closure rates at Mizuho Station are higher than those measured in bore holes at Byrd Station and in the Northern Hemisphere ice caps (Paterson 1977).

Characteristic growth processes were investigated by measuring cross-sectional areas of ice crystals for four 30 m snow cores drilled in Mizuho Plateau, Antarctica. Considerable difference was found in the growth rate of crystals between a temperature-gradient layer above 6 m depth and an isothermal layer below 10 m depth: the growth rate in the temperature-gradient layer was much larger than that in the isothermal layer. In the isothermal layer, temperature dependence of the growth rate K was expressed by an equation K = K0 exp(−E/RT), where R and T are the gas constant and absolute temperature respectively. The apparent activation energy E is 44.7 kJ mol−1.

On the other hand, in the temperature-gradient layer, the apparent activation energy was as small as 12 kJ mol−1: the difference was explained as due to the temperature gradient. Using the temperature profiles in snow that have been estimated from the meteorological data from several stations, the growth rates in the temperature-gradient layer were calculated. The calculated temperature dependence of the growth rates, taking into consideration vertical flux of water vapor between ice particles caused by the temperature gradient, showed good agreement with measured results. It is concluded that the growth process in the layer above 6 m depth is mainly due to vapor transport under the vertical temperature gradient.

Landsat-5 Thematic Mapper (TM) data from Dronning [Queen] Maud Land, Antarctica, have been analysed to provide insights into physical properties of the ice sheet. Brightness (at-satellite) temperatures calculated from digital numbers of the thermal band (TM band 6), using state-of-the-art equations and constants (Markham and Barker 1986), are 8° to 20°C lower than temperatures measured on the surface. Such differences cannot be ascribed to atmospheric absorption or to emissivity variations; instead, they suggest errors in either the sensor or instrument calibration for the temperature range 0° to −20°C. The results indicate an excess gain factor of about 1.63. The TM band 6 data seem to reproduce faithfully the relative temperature variations across the surface.

Principal components (PCs) calculated for TM bands 2, 3, 4, 5, and 7 are presented for a region of the interior ice sheet. PCs are superior to individual bands or combinations of bands for analysing the main characteristics of an image, such as rendition of surface topography (mostly given in PCl), and for distinguishing between different surface materials (mostly given in PCs 2 and 3). Band ratioing was most useful for enhancing subtle albedo variations, such as those caused by changes in surface properties due to melting.

TM data of the region, collected 18 d apart, show large changes on the snow surface; strong patterns shown in PC2 and TM bands 5 and 7 of the earlier image were nearly faded 18 d later. These strong patterns are not present at visual wavelengths or in the thermal band, and we believe they reflect wind-related variations in the physical properties of the snow.