James Ross Island (mean diameter 50 km) is located near the north-eastern coast of the Antarctic Peninsula. An ice cap, covering nearly the entire island, rises to a height of ~1 600 m. Three summer expeditions with glaciological purposes were recently achieved on this ice cap by the Instituto Antártico Argentino, two of them with the scientific participation of the Laboratoire de Glaciologie et Geophysique de I'Environnement, Grenoble.

We present results of climatic and chemical investigations performed on recent snow layers dating back about 25 a. The studied samples were collected at different sites on the upper part of the ice dome. Detailed measurements (deuterium, oxygen 18 and total β activity) were performed on more than 1000 selected samples. The relationship between stable isotope and mean annual temperature fits very well with the one previously obtained in the Antarctic Peninsula.

An ice core 22 m deep drilled on Dome Dalinqer (elevation 1600 m, mean annual temperature -15˚C) showed well-preserved seasonal variations in deuterium all along the profile, thus providing a yearly dating of the samples which was confirmed by β activity reference levels. The mean annual accumulation thus deduced is 500 kg m−2 between 1955 and 1979, with values significantly lower (30%) in the 1955–65 decade than in 1965–79. The same trend earlier observed in east and central parts of Antarctica thus appears to have a very large geographical extent.

This well-dated core allows us to undertake a year-to-year comparison between isotopic and climatological data over the 1953–78 period. The mean annual values of the deuterium content are well correlated with the average surface temperature taken over the whole Antarctic Peninsula (δD = (3.4±2.0)T - (98±32))

These data and the experimentally derived δD/δ180 relationship obtained on James Ross Island allow us to deduce a δ180 temperature gradient of 0.44‰°C−1. This low value is discussed in view of a new isotopic model taking into account the partial removal of precipitation and the possible variation of the oceanic source. James Ross Island thus appears suitable as a potential site for reconstructing past climatic changes of the Antarctic Peninsula beyond existing data.

Contamination-free techniques were used for sampling and analysing the snow samples. Na, K, Ca, and Al (by atomic absorption), H+ (by titrimetric measurements), SO4 2- and NO3 − (by ion chromatography), and conductivity were determined on more than 100 samples collected in a 4.3 m deep pit. Some of these parameters were also measured on ice-core samples or additional pit samples.

Snow impurities are contributed by different aerosol sources: sea salt, continental particles and the small-size particles produced by the conversion of various atmospheric gases. The relative importance of these sources has been estimated.

James Ross snow was found always to be slightly acid (1 to 10 μEquiv. l−1 of H+, mainly as sulphuric acid). Nitrate concentrations are much smaller (0.4 μEquiv. l−1 ). Strong seasonal variations are observed for H2SO4 deposition, probably in relation to its formation in the Antarctic atmosphere.

Sea-salt deposition exhibits also seasonal variations which can be correlated with storm frequency in the Weddell Sea area. The continental aerosol contribution is weak as indicated by very low Al values.

The influence of Deception Island volcanism on the regional aerosol chemistry is examined. A marked increase of snow acidity was detected after the 1967 eruption of this volcano, but no ash layers were observed.

The strong variations of the conductivity of melt water are interpreted: it is shown that this parameter is not representative of the extent of sea ice

Using instrumental neutron activation analysis, we have determined the concentrations of Na, CI, Al, Sc, V, Cr, Mn, Co, Fe, Zn, As, Se, Br, Sb, and J in ultra-clean dated snow samples collected in central East Antarctica. These samples had been previously analysed by flameless atomic absorption spectrometry (Boutron 1980). For the elements determined by both techniques (Na, Al, Mn, Fe, and Zn) the results are in remarkable agreement. These new data confirm and extend the conclusion that background concentrations of trace elements observed at present in Antarctic snow are close to those observed about 100 a BP and thus that Antarctic aerosol is still little affected by global atmospheric pollution. These data also confirm that, for several metals, episodes of anomalously high concentrations are in phase with increased sulphate and acidity, which could result from volcanic fallout. However, during such episodes, a straightforward relationship between the rate of increase in trace metals and the rate of increase in sulphate and acidity is not observed. This also affects volatile elements such as Cl, Br, Se, and I. This is probably due to a combined effect of local versus long-range transport of volcanic aerosols together with a physico-chemical segregation between the various components of this aerosol.

Hourly measurements of wind, temperature, and humidity were made between elevations of 1 and 20 m in the coastal area of Adélie Land over a period of 45 d in January and February 1978. About 1000 profiles are available.

The height of the constant flux layer is >5 m despite the influence of katabatic winds, the wind profile is logarithmic with a mean standard deviation from the logarithmic law of 15 mm s−1. The potential temperature profile is also logarithmic with a mean standard deviation of 0.04°C. The most important deviation from the logarithmic law appears between -4 and 0°C.

Humidity was measured at two points only, between 1 and 20 m; the mean gradient was about 0.01 mbar m−1 and the flux of latent heat was generally negligible.

Net radiation was measured for only 10 d because of radiometer failure. All the fluxes are calculated using data for the 5 m layer. For the 10 d period, the heat loss was about 3 Wm−2 with a mean heat flux of -27.5 Wm−2 and mean net radiation of -10.5 Wm−2(the minus sign signifying the flux is towards the surface). In general, the heat flux is towards the surface (92% of the cases) but is away from the surface in the afternoon. The ablation during the 10 d period without snow-fall was about 200 mm of ice.

The effect of pressurized sub-glacial water on the sliding process is quantified by calculating a “bed separation index”. The water pressure distribution i s calculated assuming the existence of a Rothlisberger channel at the bed. Kamb's formulation is used to describe the variation of normal stress over periodic bed undulations. The hypothesis is that as either basal shear stress or water pressure is increased the extent of ice-bedrock separation (on the down-glacier side of undulations) increases and enhanced sliding occurs

Data from three glaciers of widely varying size are used to test this hypothesis. For Columbia Glacier and “Ice Stream B” the importance of including the effects of water pressure in any “sliding law” are pronounced. More complete data from the third test case, Variegated Glacier, are used to compare a number of possible formulations of sliding law which encompass the above hypothesis. A modified Weertmantype law appears to be most preferable while while some possibilities, including Budd's lubrication factor hypothesis, are tentatively rejected.

Consideration of the temporal variations of the “bed separation index” reemphasize that, especially in the short time scale, variations of water pressure can dominate the sliding process. An order of magnitude increase in water discharge causes a hundred fold transient increase in the water pressure.

This paper has been accepted for publication in the Journal of Glaciology

Results of the Ross Ice Shelf Geophysical and Glaciologlcal Survey (RIGGS) provide the most complete data set available for any large portion of the polar ice sheets. In this paper, we use RIGGS data to calculate some ice-shelf characteristics. These include steady-state particle trajectories through the ice shelf and the depth of isochronous surfaces, which are of particular importance in choosing a drilling site where ice from the grounded West Antarctic ice sheet is likely to be near the surface. Our estimates for depth to ice originating from the 500 m elevation contour show good agreement with depths to a glaciochemical transition in four ice cores that is believed to be associated with this elevation. This suggests that, for much of the ice shelf, there have been no dramatic and sustained departures from steady state during the past 1.5 to 2.5 ka. With the RIGGS data and an assumed bottom melting rate distribution we calculate steady-state temperature profiles at each of the measurement stations. Then, adopting an ice-flow law deduced from laboratory experiments and ice-shelf measurements, we obtain an effective flow-law parameter for each of these sites. Using these values, the measured strain rate field is transformed to an equivalent stress field over the ice shelf. The stresses are determined by the ice-shelf freeboard and by the force field exerted on the ice shelf by its sides and by ice rises, and our analysis yields estimates of these restraining forces F for the Ross Ice Shelf. An apparent increase in F very close to the ice front suggests that the ice shelf possesses a narrow seaward fringe of anomalously stiff ice. We suspect that this represents the effects of increased bottom-melting rates (and therefore colder and stiffer ice) very close to the ice front.

In order to illustrate the role of the restraining forces in controlling ice-shelf behavior, we calculate the strain-rate field for an unrestricted Ross Ice Shelf, i.e. one that is detached from its sides and contains no ice rises. Currently the creepthinning rate for most of the ice shelf is 0.5 to 1m a−1 for an unrestricted ice shelf, but it would increase to 1 to 10 m a −1, with values up to 60 ma− 1 up-stream of the ice rises

This paper has been accepted for publication in the Journal of Glaciology.

A quantitative comparison of seasonal and interannual Antarctic sea-ice coverage over the four years 1973-76 has been accomplished through the use of passive microwave imagery from the Nimbus-5 satellite. For the entire Southern Ocean both the total ice extent (area with ice concentration greater than 15%) and the actual ice area (the spatially-integrated ice concentration) have decreased over this period of 4 a, but not uniformly in all regions. From 1973 to 1976 the annual-mean value of total ice extent decreased from 13.8 × 106 km2 to 12.1 × 106 km2, yielding an average decrease of 4.0% a−1. The inter-annual difference is greatest during the spring, as the ice decays, with the decrease in the December-mean averaging 8.4% a−1, the largest of any month. The decrease in the November-mean averaged 4.5% a−1. The overall decrease was principally due to the consistent yearly decrease of ice In the Weddell Sea sector (60°W to 20°E). Other sectors show less consistency. For instance, the ice in the Ross Sea sector (130°W to 160°E) increased from 1973 to 1974 and then decreased from 1974 to 1976, and no consistent trend is apparent in the ice extent between 20°E and 160°E. The total ice extent in the Bellingshausen- Amundsen seas sector (60°W to 130°W) actually increased slightly from 1973 to 1976. The area of the open water within the ice pack behaved differently from the total ice area, Increasing each year from February to November but having no clear interannual trend. A detailed analysis of the passive microwave imagery for the Antarctic region is planned for publication in an atlas.

This model is designed to simulate the temporal response of an ice sheet along a flow line by accounting properly for the division of the driving stress Into basal-stress and longitudinal-stress gradient components. A vertical section along a central flow line is divided into both horizontal and vertical grid points. A self-consistent solution of stress and velocity is found for any specified geometry. The sliding velocity is determined from a relationship which includes the base stress and a prescribed sub glacial water pressure. In these experiments, the ice is assumed to be isothermal. For any given set of initial conditions and a specified mass-balance distribution, the model follows the temporal behavior of the flow line to a steady state. In later work, the water pressure will be calculated from the equilibrium pressure within RothHsberger conduits for the specific geometry and determined rate of meltwater production. The time-varying temperature field due to conductive and advectlve processes will be included in later versions of the model.

Greenland and Antarctic ice-sheet surface elevations have been obtained from Seasat radar altimeter data after computer retracking of the return waveforms. The height of the altimeter above the surface is determined from the measured time between transmission of radar pulses and their return. The altimeter servo-tracking circuit attempted to maintain the midpoint of the ramp of the return waveform in the center of 60 time gates, each equivalent to 0.47 m in range. Waveforms representing an average of 100 pulse returns were recorded each 0.1 s, corresponding to a distance interval of 662 m on the surface. Deviations of the midpoint of the waveform ramp from the central-gate position were caused by changes in range larger than the design limits of the servo-circuit, thereby producing errors in the height indicated by the altimeter. If the deviation was greater than about 25 gates (13 m range), the waveform ramp moved outside the time gates and the servo-tracking was temporarily interrupted. These larger deviations resulted in a loss of about 30% of the data over the ice sheets. Both surface undulations and the steeper slopes near the ice-sheet edge produced range velocities sufficient to cause interruption of altimeter tracking. Waveforms that remained within the 60 gates have been corrected by a computer curve-fitting procedure applied to each waveform.

Preliminary contour maps of surface elevation at 100 m contour intervals have been created for much of the East Antarctic ice sheet north of 72°S and the Greenland ice sheet south of 72°N. The standard deviation of the difference in elevation at 1 032 crossover points in the retracked Greenland elevation profiles is 1.9 m, which is largely due to radial errors in determination of the satellite position. Adjustment of the radial components of the orbits to minimize the crossover differences in select regions reduces the difference to 0.25 m, which is indicative of the optimum obtainable precision over the ice sheets. This precision is comparable to the value of 0.05 to 0.10 m obtained over the oceans where waveform averages of 1 s are used. The data are sufficiently dense to permit contouring at smaller intervals (2 to 10 m) only in the regions near the maximum latitudes of ±72°. Contouring at the smaller intervals illustrates the three dimensional characteristics of some of the observed undulations.

Several methods were tested for correcting slope-induced displacements, which are typical of reflection-range measurements using a wide-angle beam. The slope-induced displacement hα2/2 is about 40 m for a satellite altitude h of 800 km and a surface slope a of 10−2. In a simulation experiment, an apparent surface profile was created by computer simulation of the altimeter measurement of an actual ice-surface profile and was then corrected for slope-induced displacement. The results show that the residual error between reconstructed and actual surfaces is about 15% of the displacement. Along the sub-satellite track the data are sufficiently dense to permit such correction for along-track slope-induced displacements caused by both undulations and regional slopes, but in the other dimension the data are generally only sufficient to permit across-track correction for regional slopes.

Routine upper-air observations for the year 1972 are used to construct the atmospheric water vapor budget between the coast and the line joining the highest points of the terrain between the meridians 55°E and 110°E. This gives seasonal and annual values of precipitation for an area of 2.4 × 106 km2. Annual accumulation amounts are then estimated for the whole budget region. By assuming that the accumulation poleward of the 2 000 m elevation contour equals the 1972 precipitation, the annual areallyaveraged precipitation (and thus the accumulation) is computed for the coastal slopes below 2 000 m.

Rigorous analysis of the rawinsonde data from five coastal stations and one on the plateau is a feature of this study. A special attempt is made to resolve the key boundary-layer properties. A systematic and important bias in the record due to missed observations during storms is demonstrated. The use of estimates for these missing values, which are obtained from multilinear regression equations, increases the computed annual poleward water-vapor transport at two stations by more than 30%.

Regions of net water-vapor inflow to and outflow from the continent were found to be organized in the same fashion as the mean circulation in the lower troposphere: the budget region is under the influence of a cyclonic vortex centered offshore. There was a poleward-directed, integrated (surface to 300 mbar) eddy flux of water vapor at all points along the coast. 43% of the annual precipitation fell during the period June to August due to a particularly stormy sequence mainly at one station.

The annual accumulation between the coast and the crest of the terrain is estimated with an uncertainty of 17%, and for the coastal slopes with an uncertainty of 39%. Comparison of the multiannually averaged accumulation derived from glaciological measurements with the computed values for a single year shows agreement within the limits of error.

In conclusion, the water-vapor budget approach produces accumulation values similar to the glaciological measurements. This approach offers a viable alternative method for monitoring the seasonal and annual accumulation over large parts of East Antarctica.

The observations of the stable oxygen isotope composition δ of precipitation at the coastal Syowa station for 1974, which are reported and analyzed by Kato (1977, 1978), are reassessed on a monthly time scale. The relationship between oxygen isotope ratio and temperature is examined in detail. The mean temperature of the preceding month, rather than the temperature at the time of sample collection, has the best association with monthly averaged δ. This one-month lagged relationship suggests the strong influence of sea ice which is related to the average temperature in the same fashion. Linear regression analyses, using monthly variations of Antarctic seaice extent for 1974, as reported by Zwally and others (1979), reveal the following interrelationships which provide good support for the contention that sea-ice extent is a dominant factor for δ values in coastal Antarctic precipitation: For August 1974 the δ values are persistently less than those predicted by sea-ice area. Over the Southern Ocean, the zonal circulation in August was weak in comparison to the 1972–79 average. It is hypothesized that this anomalous circulation resulted in the source region of moisture precipitated at Syowa being located substantially farther north.

This paper will be published in full elsewhere.

Surface melting has been observed on the lower, floating half of Byrd Glacier, Antarctica, during the height of the summer ablation season, in spite of regional air temperatures consistently below 0°C. The thickness of ice in this area is about 2 500 m, but surface crevasses penetrate to a depth of about 20 m, and bottom crevasses, being water-filled, may extend all the way up to sea-level. This leaves a zone of uncrevassed ice between which is of the order of 200 m in thickness, across which lateral shear stress due to drag against fjord walls will be concentrated. Variations in the mechanical properties of ice in this zone, specifically variations in hardness due to temperature changes, will obviously have a significant effect on the dynamics of the ice stream.

A model of the rough ice surface has been constructed, in which large crevasse furrows are represented by cylindrical V-grooves. These form the upper boundary of a solid conduction region which is semiinfinite below, and whose transient temperature distribution is calculated using the finite element method. The free surface boundary condition, that of sunlight warming the rough ice surface, is calculated by the construction and solution of coupled Fredholm integral equations of the second kind; these represent the energy absorbed at a point on the V-groove surface as being due to (1) energy directly incident from the sun, if the point is not in shadow, and (2) indirect radiation reflected from the opposite wall of the V-groove. This formulation takes into account all multiple reflections of radiation between the walls of the V-groove cavity. Additionally, the reflectivity of the ice surface is not given a constant value, but is allowed to vary, increasing as the angle of incidence departs from the surface normal.

The purpose of the model is to compare temperature distributions with a rough surface to the same model with a smooth surface. Due to the many simplifications made with regard to surface heat transfer, it is imprudent to make assertions about actual temperature distributions based on the model results, but the difference between the rough and smooth model results will provide a lower bound on the actual enhancement effect of surface roughness, i.e. future, more comprehensive, modeling of energy exchange at an ice surface will be in error by at least the predicted amount if the surface is treated as if it were flat.

The major effects of the surface roughness are greater absorptive capacity and non-uniform distribution of the absorbed energy. The greater absorptive capacity of a V-groove cavity is well known from studies in radiation heat transfer. Non-uniform distribution is due to two mechanisms: (1) the cavity effect is most pronounced at the apex of the V-groove, and (2) when surface melting occurs, energy is transported in the form of latent heat of melting from wherever the melting occurs to below the apex of the V-groove where the melt water refreezes.

The possibility that lateral shear stress is concentrated in a zone only 200 m thick means that temperature perturbations due to surface roughness need only penetrate on the order of 200 m, or possibly even less, to have a significant effect on the mechanical properties of the ice, and in turn on the dynamics of the ice stream.

While carrying out photogrammetric measurements to provide surface velocities and elevations for use in studies of the equilibrium and dynamics of Byrd Glacier, I noted that comparison of elevations obtained by ground surveys in 1978-79 with US Geological Survey topographic maps made from 1960-62 aerial photography indicated a very large apparent lowering of the glacier surface in this short timeinterval. The apparent lowering varied between 50 and 150 m along a 60 km section of the glacier for which data were available (Brecher 1980).

The ground measurements were estimated to be in error by no more than 3 in but the accuracy of elevations on the maps was unknown. Because these are reconnaissance maps, however, substantial errors would not be unexpected. It was therefore necessary to obtain more accurate glacier surface elevations for 1960–62 in order to determine whether the lowering is real. Photogrammetric strip triangulations of three individual strips of photography, two taken in November 1960 and the third in February 1963, which cover the region of the greatest apparent lowering, have now been completed. The old strips were oriented to fixed points on the two “banks” of the glacier derived for this purpose from the 1978–79 photogrammetric work, thus bringing the measurements from the old and new photography into a common coordinate system.

The glacier surface elevations for 1960–62 are the same as those obtained from the 1978–79 ground survey and photogrammetry. While it is difficult to give measures of accuracy of the results since no independent data are available for comparison, internal evidence indicates that precision higher than the expected 10 m has been achieved in the measurements. It can thus be stated unambiguously that no detectable surface lowering has occurred on any of the parts of the glacier which have been investigated.

Snow and ice-core samples from a number of sites in Antarctica and Greenland have been analyzed for the major anions Cl−, NO3 −, and SO4 2- by ion chromatography. Reproducibility on adjacent core or pit samples is ±10% at the 95% confidence level. Chloride is of marine origin except following some major volcanic eruptions. Chloride concentrations decrease exponentially with increasing site elevation with a scale height of about 1.5 km. For sites of comparable elevation, Antarctic Cl− concentrations are only slightly higher than in Greenland. Sulfate concentrations, corrected for the marine aerosol contribution, show an inverse dependence on snow accumulation rate. For sites of comparable accumulation rate, Greenland concentrations exceed those in Antarctica by a factor of 2 to 3. Nitrate concentrations also decrease with increasing accumulation rate and for comparable sites Greenland NO3 − concentrations are a factor of 2 higher than in Antarctica. There is no evidence of solar modulation or supernova perturbation of Greenland NO3 − concentrations. The Byrd deep core is shown to have distinct seasonal variations in Cl− and SO4 2- that may be used for dating. In addition, the Byrd core contains volcanic signals similar to those found in Greenland. Recent Greenland snow contains about 4 times as much SO4 2- and 2 to 3 times as much NO3 − as is found in older ice due to modern fossil fuel combustion.

Uniaxial compression tests (UCT) under constant cross-head speed and simple shear tests (SST) under constant load have been carried out in a snow cave at -16°C on ice cores within at most one month of extraction from depths of 235, 504, 708, and 896 m at Dye-3, Greenland, as part of the Greenland Ice Sheet Program (GISP), in order to improve understanding of flow behavior. UCT specimens had their stress axis at 45° from the core axis and SST specimens had their shear parallel to the horizontal plane of the core, so that in both cases maximum resolved shear stress was in the horizontal plane of the ice sheet.

Fracture stresses in UCT had characteristic values for each depth corresponding to core quality. When effective strain-rate was plotted against effective shear stress on a double logarithmic plot, (1) data from the same depth fell on the same straight line for both tests, (2) for a given stress, strainrate varied with core-depth by less than half an order of magnitude, and (3) the stress exponent lay between 2 and 2.5, significantly less than the 3 to 4 reported for polycrystalline ice.

Values extrapolated from the present data agree quite well with those deduced from bore-hole tilting measurements at Byrd station, Antarctica, and Camp Century, Greenland.

Two interdependent lines of evidence contribute to an understanding of basal ice conditions within Taylor Glacier, from the thermal regime and from features of the englacial and basal debris. The thermal regime of the basal ice in Taylor Glacier suggests mechanisms for the entrainment and transportation of debris near the glacier sole.

Assuming steady-state conditions, basal temperatures over the ablation area of Taylor Glacier were calculated from the geothermal heat in flux, the nearsurface ice temperatures, ice velocities, and the ice thickness. In over 50% of the lower ablation area the basal ice may be melting.

The englacial and basal debris is incorporated by processes operating at the glacier sole. Englacial debris is incorporated by particle regelation; basal debris is incorporated by freezing block regelation. Both processes occur close to the glacier margin, where inner warm basal ice is in contact with marginal cold ice. The debris frozen to the sole here remains relatively unmodified during transport to the margin.

Both lines of evidence indicate that Taylor Glacier is, and has been, capable of erosion and entrainment of debris by temperate ice at the glacier sole. Debris deposited along the margin i s similar to drift deposited beyond the glacier, implying that any recent climatic changes have had little effect on basal ice conditions within Taylor Glacier.

The Norwegian Antarctic Research Expedition 1978–79 used the Scott Polar Research Institute Mk IV radio echo-sounding system fitted in a Bell 206B helicopter to survey 620 km of Riiser-Larsenisen and 100 km across the outer part of Stancomb-Wills Ice Stream. Observed thicknesses of Riiser-Larsenisen decrease from 700 m at the grounding line to less than 200 m at the ice front. The thickness of Bllenga ice rise varied between 200 and 450 m. The ice shelf thins towards the east, and seems there to flow obliquely to the ice front (Fig.1).

Step-like change in thickness of >150 m over 500 m horizontal distance i s observed in the central part of the ice shelf. The records also demonstrate undulations in ice thickness of 600 to 700 m wavelength and 50 m amplitude, and various types of rifts and crevasses. Internal layering is recorded at 250 to 300 m depth over Blåenga and i n the ice shelf up-stream of this ice rise.

Observed ice thicknesses on Stancomb-Wills Ice Stream range from 130 to 220 m, with no systematic decrease towards the ice front. The records include long sections of heavy scatter from densely spaced rifts and bottom crevasses. This ice stream attains velocities > 4 km a−1, and is much more active than Riiser-Larsenisen. This high activity has resulted in extensive fracturing of the ice shelf.

The radiation balance at and above a snow-covered surface is affected not only by the high general level of the surface albedo, but also by the strong spectral variation of that albedo. Furthermore, the fact that clouds and snow have highly correlated optical properties means that the radiative problem of clouds over a snow surface is a particularly difficult one.

Both for its own intrinsic interest, and because it is in many ways as clear an example of snowatmosphere radiative interactions as one is likely to find, we have chosen to make spectrally detailed model calculations of solar and long-wave radiation over Antarctica using an atmospheric radiation model of Wiscombe (1975) coupled to the recent snow reflectivity model of Wiscombe and Warren (1980). Typical clear and cloudy situations are studied. Radiative fluxes are examined particularly at the surface and the top of the atmosphere since these are the locations of most past and future measurements.

Radiation budget calculations are compared with observations at Plateau station for various sun angles and cloud conditions. The effects on the radiation balance of a sub-visible ice-crystal cloud (“clear-sky ice-crystal precipitation”), as well as observed water clouds over the snow surface are investigated.

Because the snow of the Antarctic plateau is very clean, falls throughout the summer, and never melts, we can make accurate calculations of the surface albedo using our model for fine-grained snow. The absorbed solar radiation at the surface is almost entirely in the near-infrared where snow albedo is considerably lower than its visible values.

Snow-surface albedo increases with zenith angle for all sun angles. The spectrally integrated planetary albedo is about 10% less than the surface albedo and shows the same zenith-angle dependence for high sun. But for solar zenith angles greater than about 70° the planetary albedo may show a contrary trend because the increased atmospheric absorption in the long slant path overwhelms the zenith-angle dependence of the snow albedo.

Clouds raise the spectrally integrated planetary albedo because the cloud particles are smaller, on average, than the snow grains. Clouds raise the surface albedo by absorbing near-infrared radiation and thus altering the spectral distribution of the solar radiation reaching the surface.

The Antarctic atmosphere is so dry that more solar radiation may actually be absorbed by ozone than by water vapor, even for a water-vapor saturated troposphere, in contrast to the situation elsewhere.

As the solar absorption due to ozone is enhanced by the lack of water vapor, so is the absorption due to C02. Because some of the water-vapor absorption bands overlap CO2 absorption bands, the radiation budget is more sensitive to CO2 variations in the Antarctic than elsewhere. The radiation-budget effects of possible future increases as well as the likely Pleistocene reductions of atmospheric CO2 are investigated.

This paper will be submitted in full to the Journal of Geophysical Research. This research was supported by US National Science Foundation grant ATM-80-24641. The computations were done at the National Center for Atmospheric Research.

Data from instruments placed on tabular Antarctic icebergs show that they flex, heave, and roll in response to ocean waves. One iceberg was monitored for 13 months in the Weddell Sea by an automatic data collection platform deployed by the Norwegian Antarctic Research Expedition 1978-79. The temperature sensors demonstrated burial by snow-fall during autumn and winter, but melting during the following summer led to a net annual surface mass balance of -0.1 m water equivalent and to snow and firn temperatures being raised to the melting point. Strain and tilt instruments showed flexure and tilt with main frequencies around 15 to 20 s, but short sampling periods made it difficult to conduct comprehensive statistical analysis of the data.