Introduction
Lake Fryxell (lat. 77° 37′ S., long. 163° 7′ E.) is a permanently ice-covered lake situated in lower Taylor Valley at an altitude of 22 m. No information is available for the climate of the area, but on Ross Island 80 km. to the east the mean annual temperature is approximately −17°C. with a range of approximately 45°C. (Reference David and PriestleyDavid and Priestley, 1914; Reference FlowersFlowers, 1958), and the total solar energy of wave-length >3 μ falling is 93,000 cal./cm.2yr. (Reference Thompson and MacdonaldThompson and Macdonald, 1961).
The physiographic position of Taylor Valley modifies the climate; westerly winds descending from the polar plateau are channelled by the valley, and adiabatic heating and reduction in relative humidity result. Sections of ice core obtained by drilling through the lake ice-cover were stood vertically on its surface; periodic inspection of the cross-sectional shape of these cores showed that maximum ablation occurred during periods of westerly wind.
In November 1963 the eighth Victoria University Expedition carried out a limnological investigation of Lake Fryxell. A series of ice platforms were observed, and the measured heights between successive ice platforms were interpreted to give a measure of ablation of the lake ice.
Formation and Significance of Ice Platforms
The lake ice-cover, which varies in thickness from 3.8 to 4.5 m., is being continually replaced by accretion of ice at the ice–water interface and ablation from the surface. During summer, absorption of solar energy causes the formation of water-filled interstices within the ice column. In late summer the ice column becomes permeable by interconnection of the interstices and a water table appears at depths varying between 45 and 56 cm. below the mean surface of the ice. Interstices above the water table are then drained.
Patches of wind-blown dust derived from extensive moraines surrounding the lake accumulate on the ice surface throughout the year. During summer the dust patches absorb solar energy and pass downwards to form pits similar to the dust wells described from the surface of glaciers by Reference CharlesworthCharlesworth (1957, vol. 1). After a period of growth the pits intersect the level of the late summer water table.
In late summer some pits and interstices intersect the water table and during the freezing of early winter the water table in such situations is preserved as local horizontal platforms separated from the overlying ice by an air space. With subsequent surface ablation and addition of ice at the base of the column, these platforms become elevated above the level of the late summer water table and are eventually exposed at the surface (see Fig. 1). Being protected from the wind and incident radiation by the surrounding remnant sides of pits and interstices, these platforms tend to resist ablation. Three series of platforms at different heights, representing the repetition of this process in three successive yearly cycles, are preserved on Lake Fryxell; two platforms formed successively, and preserved one above the other, are illustrated in Figure 2.
If it is assumed that the thickness of the ice cover of the lake remains approximately constant from year to year, although varying cyclicly within a year, then the height difference between successive ice platforms gives a measure of ablation of the lake ice as a whole. At the end of summer the water table in the lake ice freezes. During the winter the ice as a whole rises relative to the water-level, as a result of the accretion of ice on the bottom and the ablation of ice from the surface of the ice cover. At the beginning of the summer (November 1963) the frozen water table in the surface ice of Lake Fryxell was already 19.7 cm. above the new water table (Table I). During the summer the effect of surface ablation far outweighs bottom melting, so that from the winter of 1962 to the winter of 1963 a total rise of 40.7 cm. had occurred. This total uplift equals the thickness of ice ablated from the surface, if it is assumed that the ice had the same thickness at the start of each winter. The assumed constancy of thickness is supported by the regularity of height differences between the three successive ice platforms and by a measurement of ice thickness in 1961–62 (Reference AnginoAngino and others, 1962), compared with that measured in November 1963. Measured heights of the 1963 ice platforms above the water table and the height differences between the 1963, 1962 and 1961 ice platforms are given in Table I.
Conclusions
The results yield a mean net ablation figure of 40.7 cm. of ice for the year-long cycle terminating with the winter freeze of 1963. The mean value for the previous cycle was 31 cm. of ice. Since no precipitation figures are available for this area, gross ablation cannot be calculated, but net ablation figures are more useful for calculating glacier ice budgets. These figures are considerably higher than those given for net ablation at similar altitudes on Ross Island 80 km. to the east (Reference David and PriestleyDavid and Priestley, 1914). Estimates based on ablation measurements for most of 1903 and 1908 are given as 15.2 and 17.8 cm., respectively. The difference between ablation values for opposing sides of McMurdo Sound is attributed to the prevalence of katabatic winds in Taylor Valley.
Application to the Ice Budgets of Canada and Commonwealth Glaciers
At opposing ends of Lake Fryxell are Canada and Commonwealth Glaciers (Fig. 3); fed by firn fields situated at altitudes between 400 and 800 m. in the Asgard Range, the glaciers descend through constricted valleys to produce broad tongues on reaching the floor of Taylor Valley. The position of the snow line divides the glaciers into regions of net accumulation and net ablation. By extrapolating the net annual ablation rate measured on Lake Fryxell to the tongues of these glaciers and by assuming the tongues are neither growing nor shrinking, estimates of ice budget and rates of ice flow can be calculated. The assumed steady-state is supported by the observations of Reference Péwé and ChurchPéwé and Church (1962) and by a comparison of ice thickness at the snouts at present with those given by Reference TaylorTaylor (1922). On Gornergletscher in Switzerland and Taku Glacier in Alaska, the increase in net ablation below the firn limit has been found to be almost proportional to the distance below the snow line (Reference RenaudRenaud, 1952; Reference NielsenNielsen, 1957). If it is assumed that the same condition exists between Lake Fryxell (where an average figure for net ablation is 36 cm.) and the snow line (where net ablation is zero), then net ablation on the snouts of Canada and Commonwealth Glaciers can be calculated by measuring the areas of the tongues between the contours at 50 m. intervals and using the interpolated ablation figures for altitudes intermediate between those of the contours.
Ice Balance for Commonwealth Glacier
If the névé of the glacier is defined as the area above the snow line and the tongue as that below, then the névé covers 46.2 km.2 and the tongue 7.7 km.2. 0.0015 km.3 of ice ablate annually from the snout, so to maintain the glacier at its present size the average net ice accumulation in its névé is 3.3 cm. annually. The narrowest part of the valley through which the ice descends is 1.15 km. wide; assuming that the thickness of the glacier in this position is 60 m., the flow rate is approximately 21 m./yr.
Ice Balance for Canada Glacier
The area of the glacier above the snow line is 23.6 km.2, while that below has an area of 6.6 km.2. 0.0012 km 3 of ice ablate annually from the snout; to maintain the glacier at its present size, the annual average net ice accumulation in the névé is 5.1 cm. The most constricted part of the valley through which the glacier descends is 1.15 km. wide; assuming that the thickness of the glacier in this position is 60 m., the flow rate is approximately 17 m./yr.
The névé of Canada Glacier lies mainly above the 1,000 m. contour, while that of Commonwealth Glacier is mainly below (Fig. 3). This height difference explains the higher precipitation in the firn fields of Canada Glacier compared with those of Commonwealth Glacier.
It is appreciated that the values given for ablation, accumulation or ice flow of the glaciers are not necessarily exact. The purpose of the calculations is to provide estimates for approximate values for this type of alpine glacier, so common in the ice-free areas of Victoria Land, in the absence of any field measurements.
Acknowledgements
Financial assistance for this research was provided by the New Zealand University Grants Committee and the Victoria University of Wellington Research Committee. Thanks are also due to the U.S. Navy for transport to and in the field, and to the Antarctic Research Division of the New Zealand D.S.I.R. for its help and hospitality.