Glaciological data collected at Patriot Hills, Antarctica (80°18'S, 81°22'W), are used to assess the local mass balance of the ice sheet. The data were collected during two field campaigns conducted by the Instituto Antartico Chileno in January and November 1995. Measurements included surveying of stakes, and ice thickness derived from discrete radar soundings with a ground-based high-frequency impulse system.

Ablation occurred on the bare-ice field at the base of Patriot Hills, with a maximum value of 7g cm −2a−1. Net accumulation was detected away from the mountains, over the firn-covered area of the glacier, with a maximum rate of 10 g cm 2 a−1 Ice thickens rapidly away from the mountains, reaching a thickness of 383 m, the maximum range of the radar system, near the center of the blue-ice field. No significant difference in surface elevation of the ice was detected over the 305 d period, which indicates that the ice is in near-equilibrium at Patriot Hills.

The variation of shear deformation rate with depth at the Dome Summit South (DSS) site, 4.7 km (~4 ice thicknesses) from the summit of Law Dome, East Antarctica, has been determined by repeated borehole inclination measurement. The results show that from the surface down to 1000m (ice-equivalent depth! deformation rates increase as expected with the increase in stress, temperature and the development of stronger ice-crystal fabrics. There is a broad maximum in strain rate around 1000 m. Below this depth, strain rates decrease, with values in the basal ice ~1/3 of those at 1000 m. in DSS, Holocene ice with low, uniform impurity levels extends to a depth of 1110 m, so the decrease in shear rale below 1000 m is attributed not to any change in properties of the ice, but to shear stress reduction induced by the large-scale retarding effect of local bedrock hills. Below 1000 m, with in the zone of retarded flow, there is a narrow spike, 14 m thick, in which the shear rate is ~5 times that in the ice immediately above and below. The shear-rate spike corresponds in depth to ice with high dust concentrations, small crystal size and strong vertical c-axis fabrics that was deposited at the Last Glaciol Maximum. A surface velocity of 1.98 ± 0.03 m a−1 obtained by integration of shear rate over the borehole depth is in agreement with the value of 2.04 ± 0.11m a−1 obtained by the global positioning system. The ratio of average column velocity to surface velocity determined by the borehole measurements is 0.74. A value of 0.76 is obtained from mass-balance considerations.

The determination of catchment boundaries is a major source of uncertainty in net balance studies on large ice sheets. Here, a method for defining a catchment boundary is developed using new measurements of ice-surface velocity and elevation near the Ice Stream B/C boundary in West Antarctica. An objective method for estimating confidence in the catchment boundary is proposed. Using elevation data, the resulting mean standard deviation in boundary location is 13 km in position or 6000 km2 in area. Applying a similar uncertainty to both sides of the Ice Stream Β catchment results in a catchment-area uncertainty of 9%. Much larger uncertainties arise when the method is applied to velocity data. The uncertainty in both cases is primarily determined by the density of field measurements and is proportionally similar for larger catchment basins. Differences in the position of the velocity-determined boundary and the elevation-determined boundary probably result from data sampling. The boundary positions determined here do not support the hypothesis that Ice Stream Β captured parts of the Ice Stream C catchment.

Flow stripes along the surface of rapidly moving ice streams are shown to be an expected reaction of a viscous medium flowing over an irregular bed, whenever the velocity at the bed is large compared to shearing through the thickness. The principal features of the process are as follows. At high basal speeds the ice acts as a strongly selective band-pass filter transmitting basal undulations on spatial scales of a few ice thicknesses very effectively to the surface. The decay of the short-scale features by diffusion is strongly retarded by horizontal stress gradients. Consequently, localized disturbances at the bed produce topographic effects on the surface that are advected long distances downstream before decaying away. This mechanism may explain many of the flow stripes on active ice streams moving by rapid basal motion.

The future behaviour of the Antarctic ice sheet depends to some extent on its current state of balance and its past history. The past history is primarily influenced by global climate changes, with some small amount of local feedback, and by sea-level changes generated primarily by the Northern Hemisphere ice-sheet changes, again with a small amount of feedback from the Antarctic ice sheet. An ice-sheet model which includes ice shelves has been used to model the Antarctic region and the whole Northern Hemisphere high-latitude region through the last ice-age cycle. For the climate forcing, the results from the global energy-balance model of Budd and Rayner (1990) are used. These are based on the Earth's orbital radiation changes with ice-sheet albedo feedback. Additional sensitivity studies are carried out for the amplitudes of the derived temperature changes and for changes in precipitation over the ice-sheets. For the Antarctic snow-accumulation changes, the results from the Voslok ice core are used with proportional changes over the rest of the ice sheet. For the sea-level variations, the results generated by the Northern Hemisphere ice-sheet changes provide the primary forcing, but account is also taken of the feedback effects from bed response under changing ice and ocean loading and from the Antarctic changes.

The results of the modelling provide a wide range of features for comparison with observations, such as the margins of maximum ice extent. For the Northern Hemisphere the results indicate that the peak mean temperature shift required for the ice-edge region is about -12°C, whereas outside the ice-sheet region this change is smaller but over the ice sheets it is larger. For the Antarctic region during the ice age the interior region decreases in thickness, due to lower accumulation, while the grounding-edge region expands and thickens due to the sea-level lowering. As a result, the derived present state of balance shows a positive region over most of inland East Antarctica, whereas coastal regions tend to be nearer to balance, with some slightly negative regions around some of the large ice shelves and coastal ice streams which are still adjusting slowly to the post-ice-age changes of sea level and accumulation rates.

The primary effects of global warming on the Antarctic ice sheet can involve increases in surface melt for limited areas at lower elevations, increases in net accumulation, and increased basal melting under floating ice. For moderate global wanning, resulting in ocean temperature increases of a few °C, the large- increase in basal melting can become the dominant factor in the long-term response of the ice sheet. The results from ice-sheet modelling show that the increased basal melt rates lead to a reduction of the ice shelves, increased strain rates and flow at the grounding lines, then thinning and floating of the marine ice sheets, with consequential further basal melting.

The mass loss from basal melting is counteracted to some extent by the increased accumulation, but in the long term the area of ice cover decreases, particularly in West Antarctica, and the mass loss can dominate. The ice-sheet ice-shelf model of Budd and others (1994) with 20 km resolution has been modified and used to carry out a number of sensitivity studies of the long-term response of the ice sheet to prescribed amounts of global warming. The changes in the ice sheet are computed out to near-equilibrium, but most of the changes take place with in the first lew thousand years. For a global mean temperature increase of 3°C with an ice-shelf basal melt rate of 5 m a−1 the ice shelves disappear with in the first few hundred years, and the marine-based parts of the ice sheet thin and retreat. By 2000 years the West Antarctic region is reduced to a number of small, isolated ice caps based on the bedrock regions which are near or above sea level. This allows the warmer surface ocean water to circulate through the archipelago in summer, causing a large change to the local climate of the region.

We present simulations of the Northern Hemisphere land ice through the last glacial-interglacial cycle with a vertically integrated ice-sheet model and a three-dimensional thermomechanical ice-sheet model. Both models are coupled asynchronously to the zonally averaged Louvain-la-Neuve climate model, which includes simplified treatments of the atmosphere, ocean and sea ice. The two-dimensional vertically integrated ice-sheet model, which contains no thermomechanical coupling, was developed in spherical coordinates (Marsiat, 1994). The three-dimensional thermomechanical ice-sheet model was developed using the two-dimensional vertically integrated model as source.

We compare results of the vertically integrated with those of the thermomechanical ice-sheet model. in the thermomechanical model the deformation properties of ice depend on the temperature within the ice and the enhancement factor; the latter is introduced to model, in a simplified approach, the different flow properties of Pleistocene and Holocene ice due to varying dust content. The computations with the thermomechanical model show that the growth and decay of the Northern Hemisphere ice sheets can be modelled with a common enhancement factor for all ice sheets. It is shown that there are model set-ups for the thermomechanical model yielding temporal developments of the total ice volume comparable to those of the vertically integrated model. Furthermore, we demonstrate that for the coupled climate/cryosphere system the total ice volume depends considerably on the enhancement factor.

A topographic map of a 120 km by 20 km section of the Amery Ice Shelf, East Antarctica, mapped with the global positioning system (GPS) in the spring of 1995, revealed two long, shallow troughs in the ice-shelf surface. Smooth features coinciding with these troughs appeared in a synthetic aperture radar image acquired 18 months earlier. ERS-1 altimeter waveform sequences and backscatter measurements along repeat satellite ground tracks across the same section of the Amery Ice Shelf, for the 1993-94 summer, exhibited a dramatic change over a 2 km sector between 30 January and 2 February 1994. The change is consistent with the presence of liquid water on the ice-shelf surface, located in the deeper of the two troughs. A time series of special sensor microwave/ imager brightness temperatures over the Lambert Glacier-Amery Ice Shelf region for the same period has sharp maxima on 5 January and 21 January 1994. These maxima are interpreted as the melting events leading to the meltstream observed in the altimeter data 25 days later.

Comparison between numerical model ice-shelf flow simulations and synthetic aperture radar (SAR) interferograms is used to study ice-flow dynamics at the Hemmen Ice Rise (HIR) and Lassiter Coast (LC) corners of the iceberg-calving front of the Filchnei—Ronne Ice Shelf, Antarctica. The interferograms are constructed from SAR images provided by the European Space Agency's remote-sensing satellites (ERS-1/2). Narrow bands of large shear strain rate are observed along the boundaries between fast-flowing ice-shelf ice and no-flow boundaries. Large rifts, opened where the ice shelf separates from the coast, appear to be filled with a melange of sea ice, ice-shelf fragments, and snow. Trial and error is used to find the best match between artificial interferograms, constructed from modelled ice flow, and the observed interferograms. We find that at both HIR and LC, ice with in the coastal boundary layers must be significantly softer than adjacent ice. At HIR the rift-filling ice melange transmits stress from one ice-shelf fragment to another; thus it must have mechanical competence and must moderate both separation of the ice shelf from the coast and the release of icebergs. However, the ice melange along the LC does not. The difference may be related to melange thickness, which could vary in the two locations due to differences in sub-ice-shelf oceanography or perhaps to regional atmospheric warming, currently under way along the Antarctic Peninsula. Future warming could weaken the melange ice around HIR as well, causing the ice shelf to lose contact with that shelf-front anchor.

An approximation common to vertically integrated and three-dimensional models used to simulate growth and retreat of ice sheets in response to changes in climate and sea level is that the component of the mean column velocity attributable to longitudinal strain rates is small relative to the component attributable to shear strain rates parallel to the gcoid, and can be neglected. We investigate the internal consistency of this sliallow-ice approximation by using a three-dimensional model of the Antarctic ice sheet to calculate the ratio of the horizontal shear strain rates, averaged through the lowest 10% of the ice depth, to the horizontal longitudinal strain rates, averaged through the total ice depth. This ratio is plotted for the regions of the ice sheet characterized by “inland ice-sheet flow”, i.e. where the surface elevations are above 1500 m in East Antarctica, and above 950 m in West Antarctica. These areas are generally inland of the ice streams. The areas where this ratio exceeds 10, 50 and 100 are delineated for the cases of (i) no basal motion, and (ii) “moderate” basal motion. The results generally support the validity of the shallow-ice approximation throughout most areas of the inland Antarctic ice sheet, although it breaks down, as expected, near flow divides and mountain ranges. The model is also employed to simulate the response of the ice sheet to linearly increasing accumula-tion rates, such that the present accumulation is doubled after 100 years and then held constant, using the surface kinematic equation. The results suggest that the accumulation increase in East Antarctica during recen t decades observed in widely spaced ice cores may be sufficient to account for a positive change in the time rate-of-change of the surface heights of up to 0.1 m a−1 during this time period, throughout the areas north of 72° S, if the accumulation increase has been widespread.

Assessment of the effect of the Antarctic ice sheet on sea level requires an accurate determination of its current state of balance. This is usually done by comparing the ice net accumulation for an area with the net outward flow of the ice. To obtain the net outward flux to better than 20% accuracy the relationship between the column-integrated ice flux and measurements of ice-sheet thickness and surface velocity must be considered. That relationship is summarised by the ratio of depth-averaged velocity ϒ to surface velocity γs in areas where ice sliding can be neglected, this ratio strongly reflects the rheological properties of the ice and depends on the shear strain-rate profile through the ice column, which is influenced by profiles of stress, temperature and ice-crystal fabric.

We present calculations from a flowline model of a hypothetical large ice sheet to demonstrate how development of an anisotropic ice-crystal fabric can influence the depth profile of horizontal velocity. The temperature dependence of ice rheology, combined with typical ice-sheet temperature profiles, yields a more block-like velocity profile with increasing to values in the range 0.89-0.96 from the isothermal rheology value of 0.8. Our results show that the onset of enhanced shear flow with increasing shear strain, as a consequence of anisotropy, and then a reduction of enhancement nearer the bedrock, can modify the velocity profile, giving values typically in the range 0.86-0.91. Smaller values than these may also occur for rougher bedrock or thinner ice.

The three-dimensional polythermal ice-sheet model SICOPOLIS is applied to the entire Antarctic ice sheet in support of the European Project for Ice Coring in Antartica (EPICA). in this study, we focus on the deep ice core to be drilled in Dronning Maud Land (Atlantic sector of East Antarctica) as part of EPICA. It has not yel been decided where the exact drill-site will be situated. Our objective is to support EPICA during its planning phase as well as during the actual drilling process.

We discuss a transient simulation with a climate forcing derived from the Vostok ice core and the SPECMAP sea-level record. This simulation shows the range of accumulation, basal temperature, age and shear deformation to be expected in the region of Dronning Maud Land. Based on these results, a possible coring position is proposed, and the distribution of temperature, age, horizontal velocity and shear deformation is shown for this column.

A linearized perturbation about a two-dimensional Vialov-Nyc ice-shect profile is used to investigate the sensitivity of the divide position at Siple Dome, West Antarctica, to small changes in the accumulation pattern and in the elevation of its lateral boundaries at the margins of Ice Streams C and D. Relaxation time-scales for the ice-sheet surface and divide position are derived from the perturbation theory. For Siple Dome, these time-scales are short: 450 800 years for surface adjustment, and 200-350 years for divide position adjustment. These short time-scales indicate that Siple Dome responds quickly to forcing at its boundaries. Therefore, the recent migration of the Siple Dome divide (determined from previous work) is probably a response to an ongoing, sustained forcing rather than a response to a long-past climate event such as the transition from the Last Glacial Maximum to the Holocene.

Based on our analysis, the inferred rate of migration of the Siple Dome divide could be attained by: (1) a steady increase in the south north spatial accumulation gradient of 0.1-1.5 × 10−9 a −2, or (2) a steady increase (decrease) in elevation of the Siple Dome lateral boundary adjacent to a relict margin of Ice Stream D (Ice Stream C) of 0.005-0.040 m a−1 over the past several thousand years. The required forcing is quite small, and suggests that major changes in the configuration of Ice Streams C and D associated with major changes in the elevation at boundaries of Siple Dome have not occurred over the past several thousand years.

Measurements of oxygen isotope ratio , major anions and cations, insoluble dust and tritium were performed every 4-6 cm along the Hercules Névé (northern Victoria Land, Antarctica) 22 m firn core. Concentration/depth profiles for H2O2, methane sulphonic acid and non-sea-salt sulphate (nssSO42-) were used to obtain a dating for the core by a multiparametric method involving a normalisation procedure and a linear combination of original profiles. This dating was compared with the and dust records to obtain a reliable identification of successive annual snow layers. The time-scale obtained from the seasonally varying signals was confirmed by an absolute date obtained from the 1965 thermonuclear atmospheric bomb test tritium peak. Around 70 years (1926-94) of annual accumulation rate data were obtained from the core. variations recorded along the core may be ascribed to seasonal variations of this parameter related to temperature variations.

Antarctic precipitation estimations derived from several new sources are examined in comparison to results found previously. The availability of analyzed atmospheric datasets has been a significant and beneficial tool for atmospheric and climate research for a broad range of research interests. This is particularly true for the polar regions, where the observational arrays are sparsely distributed. in high southern latitudes, a comprehensive assimilation of all available observations, including satellite data, is necessary for an accurate depiction of the atmospheric circulation. Recent st udies have found the operational analyses of the European Centre for Medium-range Weather Forecasts to be superior to those of other weather-forecasting centers in depicting the large-scale atmospheric circulation patterns over Antarctica. “Re-analysis” programs at major weather-forecasting centers have produced atmospheric numerical analyses using a “frozen” data-assimilation system. These projects have also derived precipitation and evaporation fields using an ensemble of short-term forecasts. From these new sources, Antarctic Ρ - E (precipitation minus evaporation/sublimation) is compared and evaluated against the long-term glaciological synthesis, as well as results from previous studies. The comparisons indicate significant regional disagreements exist between P — E from the re-analysis forecasts and the glaciological data. For the ensemble forecasting method, the continental-average evaporation is the largest area of uncertainty and differs by an order of magnitude between the rc-analysis datasets. This finding supports the use of the atmospheric moisture budget for determining P — E collectively in atmospheric diagnostic studies for Antarctica.

A model of error and variability in snow arrumulation rate is formulated to determine the reliability of accumulation-rate point measurements as regional and temporal means. The uncertainty model is applied to data from 70 shallow firn cores covering the Ross Sea drainage of the West Antarctic ice sheet. The model includes measurement error, local spatial variation and time variation. Average uncertainly in accumulation rate is 0.016maice equivalent or about 15%. Considering that measurement and depositional uncertainties are independent from core-to-core, an uncertainty of 0.01 m a−1 applies when many values are used to integrate accumulation rate over a catchment.

A mass-balance programme was initiated on Jutulstraumen ice stream, western Dronning Maud Land,Antarctica, during the austral summer 1992-93. As a part of the mass-balance programme, accumulation rate was measured along the centre line of Jutulstraumen from the shelf edge up to the plateau at about 2500 m a.s.l.

Accumulation distribution obtained from seven shallow firn cores and 48 slake readings is presented. The overall net accumulation trend displays a decreasing accumulation with increasing elevation and distance to coast, but on both the mesoscale and microscale there are significant variations. This is due to complex patterns of precipitation controlled by orography and redistribution by katabatic winds. The local accumulation distribution (few km scale) was found to be dependent on downslope surface gradient (aspect north, northwest), and variations up to 100% were found over distances of less than 3 km. The large variation in accumulation is important when selecting new core sites and for interpretation of temporal and spatial variations in accumulation derived from firn cores.

Using firn-core data from ten widely spread Antarctic sites, the dependence of firnification on temperature, wind and accumulation rate has been examined with two empirical models. One model relates the square of the porosity to the logarithm of the overburden pressure, and yields good fit to data through the first stage of firnification up to around 0.70-0.75 Mg m −3, beyond which it severely overestimates density. All three meteorological factors enter into this model, with higher temperatures and stronger winds increasing firnification rates, whilst higher accumulation rates have the opposite effect at any given depth. A temperature increase of 10°C has the equivalent effect to a wind-speed increase of 5 m s−1, or an accumulation rate decrease of 0.10 m a−1 w.e. A second model equates the logarithm of the porosity to overburden pressure and gives a much better fit to field data at higher densities where values asymptote to the bubble-free density of pure ice. This model generally yields a poor match to field data in the upper layers, with surface densities generally overestimated. Annual mean wind speed appears to be the least important of the local variables in this case, consistent with the success of the model at greater depth in matching data profiles.

The results from long-term simulations with a climate model using historical sea surface temperatures (SSTs) are used to study Antarctic accumulation rates and their relationship, or sensitivity, to temperature changes. The model used has a horizontal resolution of approximately 2° by 2°. The SST data comprise reconstructed monthly values on a global grid for the period 1950-91. The results yield an estimate for the area-averaged value accumulation rate over grounded ice of 160-180 mm w.e. a−1. The spatial pattern of the simulated trends over the period of integration indicates that the results for much of East Antarctica are consistent with evidence deduced from ice cores. The ice-sheet surface is estimated to have warmed by +0.73°C (a rate of +0.18°C per decade), while the accumulation rate is estimated to have increased by + 7.7 mm a−1 (a rate of + 1.9 mm a−1 per decade). The estimate for sensitivity is +12.5 mm a−1 per degree of warming, which can be interpreted as about -0.4 mm of sea level per year.

This paper presents model calculations of snowdrift sublimation rates for year-round automatic weather station (AWS) data in Terre Adélie, Antarctica. The model calculates vertical profiles of wind speed, temperature, humidity and suspended-snow particles in the atmospheric surface layer, and takes into account the buoyancy effects induced by the stably stratified suspended-snow profile by means of an appropriate Richardson number. The model is able to simulate accurately vertical profiles of sublimation rate derived from direct measurements. The model is used to parameterise snowdrift-sublimation rates in terms of wind speed and air temperature. This parameterisation is then used to calculate snowdrift-sublimation rates from 3 hourly data of six AWSs along a transect from Dumont d'Urville to South Pole during one year. Results show that sublimation of suspended snow is negligible in the interior of Antarctica where wind speeds and temperatures are low, whereas near the windy and relatively warm coast its contribution is significant (up to 17cmw.e. a−1). Snowdrift-sublimation rates are highest during summer, when temperatures are highest, in spite of the fact that wind speeds are not as high as in winter. It is concluded that snowdrift sublimation is one of the major terms in the surface mass balance of Antarctica, in particular in the coastal regions.