A digital elevation model (DEM) of the surface of the Antarctic ice sheet is compared with a simple two-dimensional ice-flow model to illuminate gross distortions (>500 m) of the ice-surface elevation. We use a DEM derived from ERS-1 satellite altimetry, airborne data and TWERLE balloon data. This is compared with an ice-sheet elevation model generated by applying theoretical surface elevations, calculated for two-dimensional ice flow, to isolines of distance from the grounding line (continentality). The model is scaled using only one parameter, to match the measured surface elevation at Dome Argus. The model is far from rigorous, violating continuity conditions, ignoring variations in surface mass balance and temperature, and assuming uniform basal conditions. However, the comparison of model and observed surface elevations is illuminating in terms of the behaviour of the ice sheet at a continental scale. Across the ice sheet the rms difference between modelled elevation and the DEM is around 300 m, but much of this results from isolated areas of much greater disagreement. We ascribe these gross differences to the effects of basal conditions. in four areas, the observed surface is more than 500 m higher than the modelled surface. Most of these are immediately upstream of substantial areas of rock outcrop and are caused by the damming effect of these mountain ranges. in nine areas, the measured surface is more than 500 m lower than predicted. Eight of these areas, in West Antarctica and the Lambert Glacier basin, are associated with suspected areas of basal sliding. The ninth is an area of 250 000 km2 in East Antarctica not previously noted as having unusual flow characteristics, but for which very few-data exist. We speculate that this area results from unusual basal conditions resulting in a low-profile ice sheet. A low-profile ice sheet of this size with in the East Antarctic ice sheet indicates that basal conditions are perhaps more variable than previously thought.

Radar altimeter data from Seasal (1978), Geosat (1985-88) and ERS-1 (1991—93) are employed to estimate multi-year mean changes of the surface height throughout a region on the East Antarctic ice sheet (EAIS) extending to 72.1° S, the southernmost limit of coverage for Seasat and Geosat altimetry, and above 1500 m elevation, using orbit crossover analysis. The changes are estimated on a same-season (austral late-winter (ALW) toALW) basis, where ALW is the 10 July 9 October time-frame of the Seasat altimetry. Altimeter data corrected for slope-induced errors are used. Altimeter data not corrected for slope-induced errors are also used, for comparison. Intersatellite orbit bias, combined with the effect of other radial errors such as instrumental bias, is estimated using crossover differences on the offshore ALW sea ice, which is employed as a geoid-parallcl reference surface. If similar intersatellite radial biases are characteristic of the continental Antarctic ice-sheet altimetry to 72.1° S, the results of all crossover analyses adjusted for this intersatellite bias — suggest that the mean rate-of-change of the surface height between Seasat and Geosat for ALWs 1978 to 1986-88 was with in the range +11 to -11 mm a−1. The bias-adjusted results of all crossover analyses between Seasat and ERS-1 suggest that the mean rate-of-change of the surface height between ALWs 1978 and 1991-93 was with in the range-17 to-55mma−1 (maximum intersatellitc bias estimate) or 0 to -40 mm a−1 (minimum bias estimate), suggesting that the surface may have lowered slightly during this time interval. The inconsistency of the adjusted Seasat to Geosat vs Seasat to ERS-1 results, however, may be an indication that orbits more accurate than JGM-2 are needed for estimation of regional multi-year mean changes of elevation on the EAIS. Alternatively, it may be a reflection of the differing orbit inclinations of Seasal and ERS-1.

In the spring of 1995 an extensive global positioning system (GPS) survey was carried out on the Amery Ice Shelf, East Antarctica, providing ground-truth ellipsoidal height measurements for the European remote-sensing satellite (ERS) radar altimeters. GPS- and altimeter-derived surface heights have been compared at the intersecting points of the ERS ground tracks and the GPS survey. The mean and rms height difference for all ERS-1 geodetic-phase tracks across the survey region is 0.0 + 0.1 m and 1.7 m, respectively. The spatial distribution of the height differences is highly correlated with surface topographic variations. Comparisons of GPS-derived surface-elevation profiles along ERS ground tracks show that the ERS altimeters can closely follow the GPS representation of the actual surface.

Satellite synthetic aperture radar interferometry is employed to map the hinge-line position, or limit of tidal flexing, of Rutford Ice Stream (RIS) and Carlson Inlet (CI), Antarctica, and detect its horizontal migration between 1992 and 1996. The hinge-line position is mapped using a model fit from an elastic-beam theory. The rms noise of the model fit is 1-7 mm. The hinge-line position is located with a statistical noise of 30—50 m. Using this method, we find no hinge-line migration on RIS and CI. Only the southern flank of CI, which is stagnant (velocity 10-20 m a−1 vs 100 m a−1 in the main flow of CI), retreated 376 ±36 m in 4 years. The effect of changes in ocean tide is calculated to yield a 60 m advance of the hinge-line position in our data. Hence, the detected stationarity of the hinge-line positions suggests stable ice-thickness conditions at the hinge line, except for the southern flank of CI which may be thinning. A comparison of the ice discharge calculated at the grounding line of RIS (17±2 km3 ice a−1) and of CI (2.9 ±0.3 km3a−1) with mass input from the interior regions (20 ±3 km3 ice a−1 for RIS and 2.9 ±0.4 km3 a–1 for CI) suggests a balanced mass budget for CI, whereas RIS may have a slightly positive mass budget of 3±4 km3 ice a−1.

The U.S. Geological Survey is using Landsat imagery from the early 1970s and mid- to late 1980s/early 1990s to analyze glaciological features, compile a glacier inventory, measure surface velocities of outlet glaciers, ice streams and ice shelves, determine coastline change and calculate the area and volume of iceberg calving in Antarctica.

Ice-surface velocities in Marie Byrd and Ellsworth Lands, West Antarctica, range from the fast-moving Thwaites, Pine Island, Land and DeVicq Glaciers to the slower-moving ice shelves. The average ice-front velocity during the time interval of Landsat imagery, for the faster-moving outlet glaciers, was 2.9 km a-1 for Thwaites Glacier, 2.4 km a-1 for Pine Island Glacier, 2.0 km a-1 for Land Glacier and 1.4 km a-1 for DeVicq Glacier.

Evaluation of coastal change from the early 1970s to the early 1990s shows advance of the floating ice front in some coastal areas and recession in others, with an overall small average advance in the entire coastal study area, but no major trend towards advance or retreat. Comparison of average ice-surface velocities with changes in the ice front has yielded estimates of iceberg calving. The total iceberg calving from the Marie Byrd Land and Ellsworth Land coasts during the study period was greater than 8500 km2 (estimated volume of about 2400 km-3) or an average of about 550 km3 a-1 (more than 150 km2 a-1). Almost 70% of this discharge is contributed by Thwaites and Pine Island Glaciers.

The George VI and Wilkins Ice Shelves are considered at risk of disintegration due to a regional atmospheric warming trend on the Antarctic Peninsula. Retreat of the northern margin of the George VI Ice Shelf has been observed previously, but the Wilkins Ice Shelf was thought to be stable. We investigated the positions of the northern fronts of these shelves from the literature and looked for changes on 1974 Landsat and 1992 and 1995 European remote-sensing satellite (ERS) synthetic aperture radar images. Our investigation shows that the northern George VI Ice Shelf lost a total of 906 km2 between 1974 and 1992, and an additional 87 km2 by 1995. The northern margin of the Wilkins Ice Shelf lost 796 km2 between 1990 and 1992, and another 564 km2 between 1992 and 1995. Armadas of tabular icebergs were visible in front of this shelf in the ERS images. These two ice shelves mark the southernmost documented conspicuous retreat of ice-shell margins.

Average velocities for time intervals ranging from < 1 to 15 years were measured by tracking ice-surface patterns on sequential Landsat and European Remote-sensing Satellite synthetic aperture radar images. Velocities ofThwaites Glacier range from 2.2 km a−1 above the grounding line to 3.4 km a−1 at the limit of measurements onThwaites Glacier ice tongue. The glacier increases in velocity by about 1 km a−1 where it crosses the grounding line. Over the period 1984-93, Thwaites Glacier ice tongue accelerated by about 0.6 kin a .Velocities of the floating part of several minor glaciers and some ice shelves are also determined: Land Glacier, 17—1.9 km a−1 ; DeVicq Glacier, 0.7-1.1 km a−1; Dotson Ice Shelf 0.2-0.5 km a−1; Gctz Ice Shelf, 0.2-0.8 km a−1; and Sulzberger Ice Shelf, 0.01 -0.02 km a−1. The high velocities along the Marie Byrd Land coast are consistent with the high precipitation rates over West Antarctica and, for some of the glaciers, the lack of buttressing ice shelves.

Measurements derived from remote-sensing research and Held surveys have provided new ice-velocity data for David Glacier-Drygalski Ice Tongue and Priestley and Reeves Glaciers, Antarctica. Average surface velocities were determined by tracking crevasses and other patterns moving with the ice in two sequential satellite images. Velocity measurements were made for different time intervals (1973-90, 1990-92, etc.) using images from various satellite sensors (Landsat 1 MSS, Landsat TM, SPOT XS). in a study of the dynamics of David Glacier-Drygalski IceTongue and Priestley and Reeves Glaciers, global positioning system (GPS) measurements were made between 1989 and 1994. A number of points were measured on each glacier: five points on David Glacier, three on Drygalski Ice Tongue,two on Reeves Glacier—Nansen Ice Sheet and two on Priestley Glacier. (lomparison of the results from GPS data and feature-tracking in areas close to image tie-points shows that errors in measured average velocity from the feature-tracking may be as little as +15-20 m a−1. in areas far from tie-points, such as the outer part of Drygalski Ice Tongue, comparison of the two types of measurements shows differences of about ± 70 m a−1.

Detailed surface topography is presented for two dome features of the West Antarctic ice sheet, Siple Dome and Roosevelt Island. Both these domes show linear topographic features, or “scars”, on [heir flanks. Topography is determined from a combination of existing digital elevation models (DEMs; based on satellite radar altimctry), photoclinomctry (using advanced very high-resolution radiometer images to quantitatively infer topographic details) and, in the case of Siple Dome, surface global positioning system topographic profiles. The enhanced DEMs provide heights and shapes for a variety of surface features in the vicinity of the domes, such as scars, surface undulations, ice rises and the domes themselves.

The DEMs indicate that the scar features on both domes are high relative to adjacent rapid-ice-llow areas. Scars and other related morphologic features on the flanks of both domes can be interpreted as former active ice streams and stream margins. For Siple Dome, this interpretation is confirmed by radar profiling. The evolution of the topographic height of the scars is a combination of two processes: initial elevation rise due to a positive mass balance at the shear margin in the immediate aftermath of shut-down of the ice stream; and later downslope flow as the scars become part of the dome and the dome surface velocity field. Superimposed on these events is accumulation, which buries the original shear margin but elevates the scar surface expression. Depending upon the timing of shut-flown, and the relative magnitude of these processes, the height of the scars above the current rapidly flowing surfaces may be indicative of ice-sheet thinning since shut-down, or dome expansion across former ice-stream trunks in a more or less steady-state ice sheet.

A survey of icebergs using satellite radar images has been made in the seasonal sea-ice zone of East Antarctica in the .sector between longitudes 50° and 145° E. These data provide information on the spatial distribution and size statistics of icebergs near the coast in areas not often visited by shipboard observers, and close to their sources at ice shelves and glacier tongues. The icebergs are detected and their dimensions extracted by analysis of the texture properties present in satellite images acquired with ERS-1 synthetic aperture radar during the austral winter. The minimum size of iceberg reliably detected and measured is 0.06 km2.

A significant variation, by up to a factor of two, is found in the area of icebergs close to different sections of the coast, which suggests a characteristic size for different sources. The average value of the length-to-width ratio for icebergs in the whole population shows some variability with size. The probability of finding icebergs is greatest close to the coast, decreasing in general with distance from the coast, such that few icebergs were detected more than 160 km from the coast. in one sector about 85° E, icebergs are found to at least 550 km from the coast, which is consistent with the transport of icebergs northwards in this region by a branch of the westward-heading near-coastal current (East Wind Drift) which connects with the southern margins of the eastward-heading Antarctic Circumpolar Current.

Using a three-dimensional ocean model specially adapted to the ocean cavity under the Amery Ice Shelf, we investigated the present ocean circulation and pattern of ice-shelf basal melting and freezing, the differences which would result from temperature changes in the seas adjacent to the Amery Ice Shelf, and the ramifications of these changes for the mass balance of the ice shelf. Under present conditions we estimate the net loss from the Amery Ice Shelf from excess basal melting over freezing at approximately 7.8 Gt a−1. This comprises a gross loss of 11.4 Gt a−1 at a mean rate of 0.42 m a−1, which is partially offset by freezing-on of 3.6 Gt a−1, at a mean rate of 0.19 m a−1. When the adjacent seas were assumed to warm by 1°C, we found the net melt increased to 31.6 Gt a−1, comprising 34.6 Gt a−1 of gross melt and 3.0 Gt a−1 of freezing.

Geographical changes in the height of the Antarctic ice sheet between 1978 and 1992 are mapped using altimeter data from Seasat and ERS-1 via an inverse technique that allows us to take imo account the whole altimctric error budget. in a belt between 70° and 72° S and between 150° and 80° E, a precision better than 40 cm is found for the surface elevation change, while the change in height averaged in the along-slopc direction has a precision of around 10-20 cm, leading to a precision better than 10% of the mass balance. These data suggest a relative positive imbalance of around 20% in this sector for the western and high-altitude areas, and a relative negative imbalance in some areas of lower altitude.

The areal changes of the northern Larsen Ite Shelf (LIS), Antarctic Peninsula, between March 1986 and March 1997 have been analyzed, based on synthetic aperture radar images of the European remote-sensing satellites ERS-1 and ERS-2 and on Landsat images. This analysis is complemented by data on ice motion and surface mass balance which have been obtained during several field campaigns since the early 1980s. After a period of retreat, coinciding with atmospheric warming and with decreasing net accumulation at the surface due to melt losses, the two northernmost sections of LIS disintegrated almost completely within a few days in January 1995. Recent observations of the ice-shelf section north of Jason Peninsula, which is presently the northernmost section of LIS, show increased summer melt and intensification of the rifting processes, probably causing accelerated retreat of this section in the near future. The retreat and the disintegration event of LIS indicate high sensitivity of ice shelves to prolonged perturbations of the mass balance.

The last major calving event along the Ross Ice Shelf (RIS, Antarctica) front occurred a decade ago, following a substantial increase in the rate of ice-front advance in the few years preceding the event. This “B-9” event, on the eastern part of the front between Edward VII Peninsula and Roosevelt Island, removed ≈ 5100 km2 of ice, about 100 years of advance in that sector, but reduced the ice-shelf area by only 1%. Since 1987 the entire ice front has continued to advance, more than regaining the area lost during the B-9 event. The western front is now well north of any position recorded during the last 150 years, and it lias not experienced major calving forat least 90 years. Ice-front heights generally decrease from east to west, but local variability is high. Elevations are relatively low from 171° to 177° W, the location of “warm” Modified Circumpolar Deep Water circulation beneath the outer ice shelf. Modern heights considerably exceed historical heights between 179° Wand 178° E and are lower west of 174° E, probably due to recent dynamic changes such as rifting and the western advance. The general advance of the RIS front and the period of several decades to more than a century that elapses between major calving events is consistent with a relatively stable ice front. This contrasts with several smaller ice shelves along the Antarctic Peninsula and McMurdo Ice Shelf in the Ross Sea which have retreated substantially during the past few decades.

We investigate the sensitivity of the ocean circulation in the Filchner Trough to changes in the large-scale oceanic environment and its impact on the mass balance of the Filchner Ice Shelf, Antarctica. Three experiments with a three-dimensional ocean model describe (i) the current situation, (ii) a scenario with increased ocean temperatures, and (iii) a scenario with reduced sea-ice formation rates on the adjacent continental shelf. in the final discussion brief results of a combined scenario with increased ocean temperatures and reduced sea-ice formation are presented. The changes from the current situation affect the circulation in the Filchner Trough, and melting and freezing processes beneath the ice shelf. The latter affect the amount and properties of Ice Shelf Water (ISW), a component of Antarctic Bottom Water. Net basal melt rates provide an overall measure for the changes: while the control run yields 0.35 m a−1 net melting averaged over the Filchner Ice Shelf area, the warming scenario results in a more than twofold increase in ice-shelf mass loss. Reduced production of High Salinity ShelfWater due to smaller sea-ice formation rates in the second scenario leads, on the other hand, to a decrease in basal mass loss, because the deep cavity is less well ventilated than in the control run. ISW is cooled and the ice shelf is stabilized under this scenario, which is arguably the more likely development in the southern Weddell Sea.

The King George Island ice cap, South Shetland Islands, Antarctica, was studied between 1985 and 1992. At the steady-state equilibrium-line altitude of the ice cap, the mean annual temperature is -3.6°C, the mean summer (December-February) temperature is 0°C and annual precipitation is 800 mm w.e. Precipitation increases rapidly with elevation, and annual accumulation rate at the Main Dome summit reaches 2480 mm a−1. Between 1985 and 1991 the equilibrium-line elevation averaged 140-150 m a.s.l. The ice cap has been in an overall stable state for the past 20 years, going from a weak negative to a small positive mass imbalance as increased precipitation outweighs the effects of rising temperatures. Temperatures at the bottom of the active layer over most of the accumulation area are close to 0°C, with colder temperatures down to -1.9°C in the ablation zone. Soluble impurities in the ice cap are mainly from marine sources, while undissolved mineral material amounts to only 15-54% of the total microparticle content.

The chronology of post-Last Glaciol Maximum deglaciation in the northern Antarctic Peninsula region is discussed. It is concluded that, contrary to what was earlier believed, the deglaciation process here was largely out-of-phase with that in the Northern Hemisphere. Although, for global eustatic reasons, the marine-based glaciers may have retreated simultaneously with ice-melting in the Northern Hemisphere, the land-based glaciers retreated only slowly during the first halfoftlie Holocene, about 9000-5000 BP. This may have been due either to increased precipitation counterweighing ablation or to delayed warming. A distinct but rather brief Glaciol readvancc took place around 5000 BP, probably caused by a period of renewed cooling. It was followed by the Holocene climatic optimum, about 4000-3000 BP. This warm “hypsithermal” period thus came much later than its equivalent in the Northern Hemisphere, but it roughly coincided with the Milankovitchcan Holocene insolation maximum for these southern latitudes.

Level-line surveys at a number of sites on the Antarctic Peninsula since the early 1970s have shown a lowering of the ice surface elevation in areas where the climate is warm enough for melting to occur during summer. Results are presented here from annual surveys on the ice ramp at Rothcra Point. Over an 8 year period, a large proportion of the ramp shows a generally steady reduction in surface elevation. The uppermost part of the ramp shows no clear trend. The ice ramp has suffered a mean rate ofsurfaee lowering of 0.32 ma−1 w.e. over the period of the surveys, which is similar to that seen at other sites on the Antarctic Peninsula. Measured ice velocities on the ramp are low, so the surface lowering can be attributed directly to changes in surface mass balance. The Surveys coincide with a period of long-term increase in temperature and ablation seen in meteorological records. Comparison of the observed surface lowering with temperature data shows a good agreement, and we conclude that increasing air temperatures in the region will raise ablation and increase the recession rate of the ice ramp.

The state of equilibrium of the frontal positions of the glaciers of the Dry Valleys area of McMurdo Sound, Antarctica, has heen a subject of speculation since the) were fu si seen. A programme to measure the changes in frontal positions of 12 ice-cliffed glaciers has been maintained since 1982-83, with a 30 year record for Mcscrve Glacier. Simple taped distances from fixed points to the fronts of the glaciers have demonstrated that, in proportion to their flow and ablation regimes which are two orders of magnitude slower than those of temperate glaciers, the glaciers are fluctuating as dynamically as do temperate glaciers. Maximum advance or recession rates have been found to be near ±1 m a−1 , changes which are almost undetectable in comparative photographs. Recession was found to be the dominant change trend, with yearly variations reflecting changes in annual ablation.

The total height of the ice cliff gives a ready estimation of glacier equilibrium, with cliff heights being greatest on advancing glaciers. The division between advancing and receding glaciers was found to be at a cliff height of about 25 m.

Ice-sheet thickening or thinning rates in Antarctica are measured using the “coffee-can” or “submergence velocity” method. in this, repeated measurements of the positions of firn anchors are obtained using the global positioning system (GPS). The thickness change is (lie difference between vertical velocity so obtained and long-term rate of snow accumulation. Minor corrections for firn settling and downslopc motion are made. The technique avoids difficulties of short-term fluctuations in snowfall or snow den-sification. The result for Byrd Station is near balance, -0.004 (0.022) ma−1, and for the Dragon, just outboard of Ice Stream B, thinning at -0.096 (0.044) ma−1. Uncertainties with these first results are mainly due to the short occupation times during the first GPS surveys.