Measurements of crystal size have been made on seven firn cores drilled at sites covering a range of mean annual temperature from –12.6° to –52.5°C and a range of accumulation rate from 52 to 315 kg m−2 a−1. The sorting coefficient, which gives a measure of the dispersion of crystal sizes within a sample, shows an overall pattern when data from all cores are grouped together as a function of depth. The values are generally small near the surface, increasing to a maximum around 8 m depth, then decreasing but becoming more diffuse at greater depths. Below about 5 m depth, the crystal size increases at an essentially constant rate, which depends on temperature, but in the upper 5 or 7 m the size increases at 1.5 to 2 times this rate. The seasonal variation in temperature enhances the effective mean growth rate of crystals in the near-surface layers compared to conditions with a constant mean temperature and accounts for a part of that increase. But it is likely that vapour diffusion along strong vertical temperature gradients causes the greater part of the observed increase in growth rate. The dependence of crystal-growth rate on temperature is consistent with the Arrhenius-type relation found by other studies.

Extensive radar ice-thickness sounding of ice streams A, B, and C, and the ridges between them, has been carried out. Closely spaced flight lines, as well as ties to numerous ground stations, have enabled us to compile a detailed ice-thickness map of the area. The map reveals a highly complex pattern of ice-thickness variations, which, because they are much larger than the surface relief, largely reflect the subglacial topography. Several cross-sectional profiles across the ice streams and ridges are shown, and a new configuration for Ice Stream A is presented.

Ice Stream A is connected to Reedy Glacier and Horlick Ice Stream by subglacial troughs that converge down-stream. The single trough continues, at a depth of more than 1000 m below sea-level, beneath the entire length of the ice stream and adjacent part of Ross Ice Shelf. Ridge AB (part of which may be a remanent ice stream) overlies a deep bed with pronounced troughs at its headward end; the bed shoals rapidly down-stream to a height more than 500 m above the beds of the adjacent ice streams. Ice stream B1 overlies a subglacial trough that is deep inland and also shoals markedly toward the grounding line. Near its head. Ice Stream B2 is as much as 1000 m thinner than Ice Stream Bl, but then remains much more nearly constant in thickness along its length. Ridge BC is characterized by a smoother bed and less variation in bed depth than ridge AB. Ice Stream C, which is inactive, is particularly marked by uncorrelated maxima and minima in surface and bed topography.

There are no distinct topographical steps that demarcate the transition from sheet to streaming flow at the head of the ice streams, and the ice streams are placed asymmetrically in some places with respect to their subglacial troughs. This may reflect a relative impermanence or transient behavior of the “Ross” ice streams.

Recent airborne radar sounding has made it possible to map accurately three of the West Antarctic ice streams that flow into Ross Ice Shelf. In previous work we have shown that ice streams A and Β have negative mass balances, whereas inactive Ice Stream C has a strongly positive balance. In this paper we examine in more detail the balance of ice streams A and Β by constructing several gates across them where velocities and ice thicknesses have been measured. We then examine the net fluxes in blocks of the ice streams delimited by successive pairs of gates.

Ice Stream A as a whole is apparently discharging more ice than is being accumulated in the catchment area, and currently thinning at the rate of 0.08 ± 0.03 m a−1. The situation on Ice Stream Β is more complex. We have calculated separately the fluxes from tributary ice streams Bl and B2, and examined their individual fluxes within Ice Stream Β by tracing the suture zone between them down-stream of their confluence. The flow band that is the farthest up-stream (girdle), encompassing both Ice Stream Bl and Ice Stream B2, shows a strongly negative net flux that we attribute to lateral and headward expansion of the ice streams within the band. Such expansion can occur by lateral movement of an ice-stream boundary, by temporally accelerating ice flow at the head of the ice stream, or by activation of formerly slowly moving “island” or “peninsula” ice.

The imbalance in this flow band, 8 ± 2 km3 a−1 (equivalent mean rate of change in ice thickness, is nearly half of the total excess outflow for the Ice Stream Β system (20 ± 4 km3 a−1),

— the remainder is mostly the difference between flow through the uppermost gate and mass input to the catchment area .

When for the whole of Ice Stream Β is plotted against the distance along the entire Ice Stream B, the overall pattern appears to be of mild thinning in the catchment, intense thinning in the girdle, and thickening in the main body of the ice stream, which decreases with distance from the girdle. This global behavior is suggestive of a major transient response, resulting from either a change in the internal dynamics or an internal adjustment to a change in the external forcings. We argue that there are a number of conditions which could lead to this type of response pattern. One possibility is a surge. Although the distribution of the changes in thickness is one characteristic of a surge, we caution that this alone is not sufficient to classify the behavior as a surge. Several other possibilities that support a picture of Ice Stream Β as a system in the process of dynamic change and in unsteady state are discussed.

At present, Ice Stream C and its catchment area are thickening over their entire area The present surface elevation does not suggest that Ice Stream Β has captured part of Ice Stream C. Moreover, the shut-down of Ice Stream C and the large mass imbalance of Ice Stream Β are not related.

A set of Landsat-5 multispectral scanner (MSS) images of Filchner-Ronne Ice Shelf was commissioned by the Institut für Angewandte Geodäsie and recorded over a short period in early 1986. A mosaic of digitally enhanced images was constructed, using all available ground control. In due course it is planned to use the data for the production of a series of digital-image maps at a scale of 1:1 000 000. Meanwhile a line map has been prepared at a scale of 1 :2 000 000 to show the principal glaciological features. The new mosaic shows much more detail than earlier Landsat images and extends coverage to higher latitudes. More of the poorly known southern boundary of the ice shelf can be identified but floating ice still extends beyond the limit of coverage. Extensive ice rumples are revealed, many of them undetected in earlier imagery. Flow bands can be tracked without a break from their ice-stream source to the ice front, over distances of up to 800 km. Even minor tributaries, and sheet flow from the inland ice sheet between the ice streams, yield narrow but identifiable flow bands. Ice thickness, velocity, and mass-flux data are reviewed, in order to underline their dependence on good mapping.

The horizontal divergence of drifting snow was estimated from the ice-sheet topography on Mizuho Plateau, East Antarctica. The calculation was made by using a relationship between the snow-drift transport rate and wind speed estimated from the surface slope. The divergence thus estimated for Mizuho Station (70°42′S, 44°20′E) was consistent with observations of surface net mass balance, precipitation and sublimation. Around the southern region of the Yamato Mountains, a large divergence was predicted and this is believed to be the principal cause of the bare ice field. Other factors in the formation and preservation of the bare-ice area are discussed.

A collation of available data shows that sea-water with a temperature 3°C above the in-situ freezing point lies beneath George VI Ice Shelf in the Antarctic Peninsula, and is widespread on the Amundsen-Bellingshausen continental shelf. The presence of warm water is a factor in the recent and continuing disintegration of ice shelves in the region, yet the meteorology and oceanography of the sector are little known. We discuss a plausible link between the present climatic conditions, sea-water characteristics and the warm-water intrusion on to the continental shelf, thereby illustrating an indirect climatic influence on the mass balance of ice shelves.

Detailed measurements of surface topography, ice motion, snow accumulation, and ice thickness were made in January 1974 and again in December 1984, along an 8 km stake network extending from the ice sheet, across the grounding line, and on to floating ice shelf in the mouth of slow-moving Ice Stream C, which flows into the eastern side of Ross Ice Shelf, Antarctica. During the 11 years between surveys, the grounding line retreated by approximately 300 m. This was caused by net thinning of the ice shelf, which we believe to be a response to the comparatively recent, major decrease in ice discharge from Ice Stream C. Farther inland, snow accumulation is not balanced by ice discharge, and the ice stream is growing progressively thicker.

There is evidence that the adjacent Ice Stream B has slowed significantly over the last decade, and this may be an early indication that this fast-moving ice stream is about to enter a period of stagnation similar to that of Ice Stream C. Indeed, these large ice streams flowing from West Antarctica into Ross Ice Shelf may oscillate between periods of relative stagnation and major activity. During active periods, large areas of ice shelf thicken and run aground on seabed to form extensive “ice plains” in the mouth of the ice stream. Ultimately, these become too large to be pushed seaward by the ice stream, which then slows down and enters a period of stagnation. During this period, the grounding line of the ice plain retreats, as we observe today in the mouth of Ice Stream C, because nearby ice shelf, no longer compressed by ice-stream motion, progressively thins. At the same time, water within the deformable till beneath the ice starts to freeze on to the base of the ice stream, and snow accumulation progressively increases the ice thickness. A new phase of activity would be initiated when the increasing gravity potential of the ice stream exceeds the total resistance of the shrinking ice plain and the thinning layer of deformable till at the bed. This could occur rapidly if the effects of the shrinking ice plain outweigh those of the thinning (and therefore stiffening) till. Otherwise, the till layer would finally become completely frozen, and the ice stream would have to thicken sufficiently to initiate significant heating by internal deformation, followed by basal melting and finally saturation of an adequate thickness of till; this could take some thousands of years.

Aero-measurements during the 1983–84 and 1985–86 field seasons showed that the ice in the central part of Filchner-Ronne Ice Shelf is more than 250 m thicker than has been assumed previously. In the margin area a double layering of the ice was found, with disappearing bottom reflections farther inland. High absorption of approximately 0.3 dB/m in the lower layer and a decreasing reflectivity at the ice / sea-water interface, probably caused by bottom freezing, have been estimated. Aero-measurements were used to map the surface elevation of the ice, with an accuracy of approximately 3 m. From the plot of ice thickness against surface elevation it was possible to obtain a calibration curve for isostatic conditions. Large deviations from this plot with an ice thickness which is apparently too small (they seem to be isostatic anomalies) were found in the central part of the ice shelf. The true ice thickness could easily be calculated and mapped from these anomalies and the electromagnetically measured thickness.

From the map based on flights made in 1983–84 the site for a bore hole was chosen in the central part of the ice shelf with an estimated ice thickness of about 450 m, instead of the less than 200 m deduced by an electromagnetic reflection (EMR) method. Hot-water drilling by the ground party of our group (H. Engelhardt and J. Determann) revealed an ice thickness of 465 m at this site.

In 1980–81, 1983–84 and 1985–86 airborne surveys with an electromagnetic reflection (EMR) system were made of Ekström Ice Shelf, Antarctica. The EMR data were supplemented by measurements of surface elevation with radar altimetry during flights at a constant pressure altitude. The accurate measurement of ice thickness in areas with clearly developed bottom reflectors was used to generate a plot of surface elevation against ice thickness. The effect of changing barometric pressure during the flights could be reduced by this means. Elevations were calibrated over the open sea at the beginning and end of each flight.

On the basis of these data, the surface elevation, ice thickness and isostatic anomalies have been mapped over the ice shelf.

For the study of drifting snow a pulse-counting device was developed. The detector is sensitive to the momentum transfer of individual snow particles to an exposed surface. More than 105 particles can be counted per cm2/s. The sensor is adjusted to snow particles of 30–200 μm in diameter and a velocity range of 5–30 m/s. A multi-channel analyzer system is used for data handling and to display the momentum spectrum in real time.

The mass balance of the drainage area of Ice Stream B, West Antarctica, is calculated from new measurements of both discharge and accumulation rate. The discharge is computed for a transverse section near the lower end of the ice stream. Velocities have been obtained for 787 sites, using repeated photogrammetry, with ground control by Transit (doppler) satellite tracking. Thicknesses have been obtained by radio echo-sounding. The uncertainties in the discharge calculations are only about 3%. Net accumulation is derived from profiles of gross beta activity and from identification of the 1954–55 and 1964–65 nuclear-bomb strata. The major uncertainties are associated with the identification of the catchment area and with the accumulation rate. Accumulation rate varies locally, probably due to the interaction of katabatic wind with local slope, and many spot measurements are needed to obtain a good regional mean. The integrated input is 21.4 ± 5.2 km3 a−1, and the output is 30.0 ± 1.0 km3 a−1. The deficit is thus 8.6 ± 6.2 km3 a−1, which corresponds to a mean thinning rate of 0.06 m a−1 ± 0.04 m a−1. The difference from earlier estimates is mainly due to the refined catchment area and accumulation. The imbalance is significant but smaller than previously calculated: for balance the accumulation rate or catchment area would need to be about 39% larger or the ice stream velocity would need to be 28% slower.

Analysis of an ice sample with an estimated age of 125 years, from the Antarctic Peninsula, using a scanning electron microscope with a cold stage and an X-ray micro-analysis facility, shows that H2SO4 occurs mainly at triple junctions. Sea salts show no such localization. The different behaviour may be due to the freezing-point behaviour of each chemical substance, and to the effect this has both in the atmosphere and during recrystallization in the ice sheet. If this finding applies generally to other parts of the Antarctic ice sheet, it has major implications for many of the physical properties of Antarctic ice. In particular, it leads to a better understanding of the d.c. electrical conductivity of such ice.

Detailed seismic short-refraction profiling was conducted on Ice Stream Β (UpB) during the 1983–84 austral summer. A new high-resolution data logger, developed at the University of Wisconsin, recorded both compressional- and shear-wave arrivals. We report here on P-wave and S-wave profiles recorded along a line parallel to the axis of the ice stream. Source-receiver separations up to 720 m yielded seismic velocity-depth curves to below the firn-ice transition zone (slightly greater than 30 m at UpB).

For the compressional-wave profile, geophones were separated by 2.5 m, which yielded a velocity-depth curve with a granularity of ∼1 m. The corresponding density-depth curve agrees well with direct density measurements obtained from a core extracted nearby (Alley and Bentley 1988, this volume). Discontinuities in the velocity gradient do not appear at the “critical densities” as they did at Byrd Station, Antarctica, and elsewhere (Kohnen and Bentley 1973, Robertson and Bentley 1975).

Two shear-wave profiles were recorded, both with geophone spacings of 5 m, one with longitudinal polarization (SV) and the other with transverse polarization (SH). There is a marked difference in velocity between the SH and SV waves, particularly in the shallow firn. We suggest that a strong vertical shape-and-bonding fabric in the shallow firn, as observed in cores collected at UpB, would account for this disparity.

Both in the interior of West Antarctica and on Ross Ice Shelf the ice column is dominated by ice with a distinct preferred c-axis orientation. An understanding of the dynamics of the West Antarctic ice sheet requires that we know the gross crystalline structure that characterizes each of its glaciological regimes (i.e. ice sheet, ice stream, and ice shelf). An important question is whether the strong fabric of the interior ice is preserved when this ice passes through the zone in which it is accelerated from sheet flow to stream flow, a zone that must be marked by strong longitudinal extension. Using generalized inverse techniques we have inverted seismic-reflection travel times observed at Upstream Β camp (on Ice Stream B) to obtain the gross crystalline structure of the ice column at that location. We find that the strong fabrics of the interior are indeed destroyed; only a slight preferred orientation remains. The evidence at Upstream Β camp is particularly strong because shear waves of both polarizations, which are particularly sensitive indicators of anisotropy, were analyzed as well as P-waves.

An array of nine seismographic stations, each sensitive to all three components of motion, was deployed on Ice Stream B, West Antarctica, during the austral summer of 1985–86. The network was sensitive to high-frequency (=400 Hz) seismic activity within a 350 km2 area of the ice stream, and the deployment geometry allows the precise determination of depths for events beneath the 10 km2 array. Microearthquakes from both beneath and beside the ice stream were detected and recorded (Blankenship and others 1987). Inversion of P-wave and S-wave travel times and radiation patterns allows the determination of locations and fault-plane solutions for many of these events.

We find that bottom events involve low-angle thrusting, in the down-stream direction, of ice or till; displacement is ∼½ cm per event over a (15 m)2 area. Such faulting is rare and releases an insignificant part of the total energy dissipated by ice flow. However, this is a possible mechanism for plucking of the ice-stream bed.

Fault-plane solutions for most major surface events are consistent with the opening of tensional fractures oriented transverse to ice flow. Precise location of these events shows that they correspond to open crevasses, mapped by Vornberger and Whillans (1986), that are oriented transverse to ice flow.

In addition, shear-wave splitting observed on some of the microearthquakes shows that the c-axes in the ice stream are slightly, but not strongly, anisotropic. Precise location of the sources requires the use of a detailed velocity-depth profile in the firn, which was obtained by seismic short-refraction studies (Anandakrishnan and others 1988, this volume). A density-depth profile calculated from these velocities agrees well with direct density measurements on a single core nearby (Alley and Bentley 1988, this volume).

Solar radiation incident on, and reflected by, the snow surface was measured near the South Pole as a function of wavelength, angle, and distance from the station. The objectives of the study were: (1) to observe spectral albedos of snow across the solar spectrum, (2) to obtain depth profiles of snow-grain radius in order to construct theoretical models of spectral albedo for pure snow, (3) to document the extent and degree of soot pollution due to station activities and to assess whether it could invalidate solar-radiation measurements made close to large stations, and (4) to obtain the spectral distribution of incident solar radiation at the Antarctic surface for various cloud conditions, in order to test radiation models of the Antarctic atmosphere.

Spectral albedo, measured under diffuse lighting conditions (overcast cloud) on many days, repeatedly agreed with the results of theoretical models which predicted values approaching unity in the visible and found grain-size to be the most important variable controlling snow albedo in the near-infra-red. A representative albedo curve is shown in Figure 1.The visible albedo values were found to be 98–99% and were relatively insensitive to grain-size. (These results disagree with the only previous measurements of Antarctic snow albedo which had good spectral resolution: those of Kuhn and Siogas. Their maximum albedo was only about 90% in the visible.) The near-infra-red albedo, however, varied substantially among the experiments, due to day-to-day variations in snow grain-size, caused by precipitation and wind drifting. The experimental points in the figure match theoretical calculations for grain radius less than 50 μm at wavelengths beyond 1.5 μm, and 50–100 μm for shorter wavelengths. At the shorter wavelengths the light penetrates more deeply into the snow, so the albedo is sensitive to grains beneath the surface, whereas at the longer wavelengths the albedo is influenced only by the grains very close to the surface. The observed albedos can thus be explained by an increase in grain-size with depth.

In order that our measurements would be representative of large areas, we were concerned to avoid possible effects of pollution from the station. We collected samples from the top 20 cm of snow, melted and filtered them, and analyzed the filters. The conclusion is that the pollution is very minor. Just 500 m up-wind of the station there is normally less than 1 ng of carbon per gram of snow (1 ppb). Even down-wind of the station the carbon content did not exceed 3 ppb. For snow grain-sizes typical of Antarctica, our models predict that 15 ppb carbon would reduce snow albedo by only 1% at the most sensitive wavelength. Thus we reject our earlier suggestion that the low visible albedos of Kuhn and Siogas were due to impurities in the snow and now favor other explanations.

Recent glaciological work on Ronne Ice Shelf has focused on an assumed flow line which extends from Rutford Ice Stream grounding line to the ice front. Results from doppler satellite surveying and radio echo-sounding are used in kinematic calculations to determine the basal mass balance, assuming the flow line to be in a steady state. Models suggest that basal melting dominates over most of the flow line and is most pronounced at the extremities. In the region within 300 km of the grounding line and over the final 45 km before the ice front, at least 1 m/a on average must melt away to maintain the observed velocity and thickness profile. More gentle melting occurs over about half the remaining distance, but in a region between 130 and 300 km in from the ice front, basal freezing must occur at an average rate of about 0.1 m/a to maintain a steady state. The existence of a thin layer of saline ice underlying the ice shelf, which persists for a further 80 km down-stream before being melted away entirely, is consistent with the weak returns observed during both airborne and ground-based radio echo-sounding in this region.

Glaciological field programs may be regarded as imperfect sampling schemes designed to provide fundamental physical information on the dynamics and climatic sensitivity of the Antarctic ice sheet. Uncertainty arises as a result of technical and human factors such as: (i) logistic and financial constraints, (ii) measurement errors, (iii) low spatial resolution (see (i)), and (iv) (possibly!) misconceptions on the part of glaciologists who plan and execute field work. Regardless of such uncertainty, we depend on field data as the fundamental intellectual driving force of glaciology. Introspective evaluation of our field methods and program designs is thus reasonable, and perhaps necessary, to insure that our field programs are indeed satisfying their intended purpose.

In our study, we conduct a variety of Gedankenexperimente (imaginary field programs), which sample an arbitrary, idealized ice shelf, subject to fluctuations and climatic changes on a variety of time and space scales. The “actual ” behavior of this ice shelf is produced by a time-dependent numerical simulation of ice-shelf evolution under specified forcings, using a model based on that of Lange and MacAyeal (1986). Each Gedankenexperiment consists of a spatially incomplete sampling of the model grid data at a particular moment in the evolution of the ice shelf (just as a real field program presently would sample the current state of an Antarctic ice shelf). The spatial sampling patterns are based on particular techniques commonly used in field programs (Kohnen 1985, Bindschadler and others 1987, Doake and others 1987, Shabtaie and Bentley 1987). Such sampling is designed to simulate field techniques such as airborne radio echo-sounding, surface geodetic measurements, aerial photography, and satellite altimetry (Fig. 1). We also add “random noise” to the sampled data, to simulate instrumental and navigational uncertainties.

Having sampled the idealized ice shelf by using an imaginary field program, we “process” the supposed field data in order to test how well it reveals certain aspects of ice-shelf flow and evolution. This test is conducted by comparing the field-program results with the “known” behavior (by definition) of the numerical simulation. A variety of field-program design schemes are compared on the basis of their ability to predict: (i) the long-term growth or decay of the ice shelf, (ii) the “current” state of mass balance, (iii) the “current” partitioning of ice-stream input, and (iv) the balance of forces acting on the grounding line, and the tendency of the balance to change with time.

A major aim of our study will be to point out how seriously the understanding of current ice-shelf dynamics and the ability to measure initial effects of global climatic changes (due to CO2 warming) are hampered by: (i) inability to map accurately all the regions of ice-shelf grounding, and (ii) inability to distinguish the effects of short-term variability from long-term, large-scale trends. To simulate the effects of ice-shelf grounding and ice-stream -temporal fluctuations, we specify in our idealized simulations that: (i) several ice rumples occasionally appear or disappear, and (ii) ice-stream fluxes, which feed the imaginary ice shelf, fluctuate (arbitrarily) with periods of 300 years.

Since we assess the Gedankenexperimente in terms of their ability to detect long-term climatic trends, we run the ideal ice-shelf simulation forward in time until a statistically steady state is achieved (that is, all thickness and velocity patterns are stationary when averaged over the time-scale of fluctuation). At this point, we conduct the imaginary field programs in our study. Our main intention is to determine which Gedankenexperiment can best “see through” the short-term transient “noise” of the ideal ice-shelf evolution to detect the long-term condition of steady state.

Field data is presented to support the hypothesis that Crary Ice Rise (on Ross Ice Shelf, Fig. 1) has substantially increased in area over the last 500 years, in response to ice advection through the mouth of Ice Stream B. The up-stream end of the ice rise is now surrounded by ice shelf that is currently thickening at 0.44 0.06 m/year (under an assumed zero basal melting rate). This rate of thickening suggests that the ice rise's contribution to back-stress resistance of Ice Stream B's flow, presently calculated to be 50% of the total back stress, is growing in the course of time. We speculate that this current development of the ice rise is the precursor to the possible future stagnation of Ice Stream B. It is convenient to conceptualize a possible see-saw oscillation between ice-stream surging and ice-rise build-up.

The air found in polar ice becomes isolated from the atmosphere during the transformation of snow into ice. The total gas content, V, measured on selected ice samples depends theoretically on the atmospheric pressure and temperature prevailing at the ice-formation site and also on the volume, Vc, of the firn pores at the time they become isolated from the atmosphere in newly formed ice.

In spite of scattered individual values, a clear decrease in mean v-results has previously been observed for increasing elevations at the ice-formation site. Assuming that the pores close off at atmospheric pressure, this decrease primarily reflects the pressure-elevation gradient and typically only about 10% of the decrease can be attributed to the temperature-elevation gradient or changes in Vc.

Measurements of the total gas content of Antarctic ice cores are potentially the most direct and sensitive indicator of the elevation at which the ice was formed and provide a unique tool for reconstructing the past elevation of the ice-sheet surface. Nevertheless, to make such a reconstruction with confidence, we need first to evaluate the individual contributions of the different parameters which should influence Vc. Along this line, a stage was reached with the semi-empirical observation of a quasi-linear correlation between mean Vc and temperature, although the data points on which the correlation is based are relatively scarce. Furthermore, confidence in the paleo-interpretation of V-measurements is dependent also on an understanding of the scatter in individual V-values generally observed for a given ice-core section.

New measurements of V are presented, which permit a better and more detailed description to be made of the changes in V with atmospheric pressure and elevation under approximately present-day conditions. As a first step, scattering in the values of individual V-measurements of a single core section is investigated. The experimental method used for this work is basically the same as the one used for establishing the relationship between V and elevation, as given by Raynaud and Lebel (1979). Generally measurements performed on the same horizontal layer (about 3 cm thick) reproduce the same result within the experimental uncertainty (which is ±1%). On the other hand, measurements performed on adjacent horizontal layers may show V-differences as high as 20-25%. This confirms that the variability of the V measurements that occurs over short intervals in the core is not due to experimental uncertainties and should mainly reflect changes in Vc.

The main reason for making this new set of measurements was to increase significantly the number of sites investigated in the study of the relationship between V and elevation. The work of Raynaud and Lebel (1979) was based on six sites. The results for 14 sites (in Antarctica, Greenland, and on Mount Logan in Canada) are now available, with elevations ranging from about 340 to 5300 m and temperatures in the range −12.4° to −55.4 °C. The results confirm a general decrease in Vc with temperature. But at least one site shows mean V-values which are clearly incompatible with this general trend.

Finally, before proceeding further with studies of the distribution of V versus site parameters, we need to correct our V-measurements for the loss of gas which occurs when cutting the ice samples (and consequently opening air bubbles at the sample surface). We have developed, for this purpose, a technique which calculates random distributions of bubbles in the samples and counts the cut bubbles. This calculation depends on bubble shape and size as well as on sample volume and V. In the case of the samples used in the present study, between 3 and 10% of gas is typically lost during cutting. Bubble-size measurements have yet to be made before an accurate correction can be applied to the measurements presented here.