The mass balance of a cold glacier: Meserve Glacier, south Victoria Land, Antarctica*

Transcription

1 MASS BUDGETS : REGIONAL STUDIES 429 MCLAREN, W. A A study of the local ice cap near Wilkes, Antarctica. AN ARE Scientific Report No Melbourne. Antarctic Division, Department of Supply. MÄLZER, H Das Nivellement über das Grönländische Inlandeis. EG1G , Vol. 3, No. 1. Kebenhavn. MORGAN, P. J. (Unpublished). Photogrammetric and geophysical methods of determining the mass budget of an ice cap. M.Sc. thesis presented to Department of Meteorology, University of Melbourne. OESCHGER, H., ALDER, B., LOOSLI, H. and LANGWAY, C. C Radiocarbon dating of ice. Earth and Planetry Science Vol. 1, 2, p PFITZNER, L. (Unpublished). Wilkes ice cap project ANARE Research Report, Antarctic Division, Department of Supply (in preparation). RADOK, U., JENSSEN, M J. D. and BUDD, W. F Steady state temperature profiles in ice sheets. JSAGE Symposium. ROBINSON, E. S On the relation of ice surface topography to bed topography on the South Polar Plateau. /. Glaciol., Vol. 6, No. 43, p SHUMSKY, P. A On the therory of glacier variations. WGG-IASH Bulletin VIII, Année No. 1, p WALKER, D. J Wilkes geophysical surveys, Antarctica Record 1966/129. Canberra. Bureau of Mineral Resources Geology and Geophysics, Dept. National Development. The mass balance of a cold glacier: Meserve Glacier, south Victoria Land, Antarctica* BY C. BULL and C. R. CARNEIN Institute of Polar Studies and Department of Geology, The Ohio State University, Columbus, Ohio ABSTRACT The mass balance of Meserve Glacier, on the south side of Wright Valley, Antarctica, is nearly in equilibrium. During the year November 1965-November 1966, the 9-9 km 2 glacier lost 60 X 10 6 kg of ice, equivalent of 0-61 g cm" 2 over the whole glacier. In that year the snow accumulation was less than usual; in "normal" years the balance may be slightly positive. On the 1-8 km 2 ablation tongue of the glacier the annual loss ranges up to about 34 g cm" 2, near the snout at 440 m elevation. Of this amount, 40 per cent occurs during the cold months from mid-february to mid-november. During the summer, a small amount of melting occurs on the 15 m-high cliffs around the glacier tongue, and around the margins of the glacier's upper surface, but only 2 or 3 per cent of the total loss is by meltwater runoff. Dry calving from the cliffs, which are approximately in equilibrium, amounts to about 1-5 per cent of the total mass loss from the glacier. Evaporation from a melting surface may account for 40 per cent of the total loss. The remaining loss is by sublimation. Contribution No. 124 of the Institute of Polar Studies, The Ohio State University, Columbus, Ohio 43210

2 430 ISAGE FIG. 1. Map of Wright Valley area of south Victoria Land, Antarctica, simplified from New Zealand Antarctica 1:250,000 Reconnaissance Series, Ross Island and Taylor Glacier Sheets.

3 MASS BUDGETS : REGIONAL STUDIES 431 Introduction Meserve Glacier (77 35'S, 'E) is one of a series of small glaciers on the south side of Wright Valley, south Victoria Land, Antarctica (Fig. 1). It is about 7-2 km long, with a total area of 9-9 km 2. The accumulation zone (area 8-1 km 2 ) is a basin in the Asgard Range, between 1200 and 1500 m elevation. The general balance line is in a crevassed area at about 1130 m, on the shoulder of Asgard Range, and from here the glacier extends in a relatively uncrevassed tongue (area 1-8 km 2 ), about 600 m CONTOUR NTtP\W- 50 MFrtRS VELOCITY VECTOR A CONTROL POINT -< PHOTO CONTROL POINT CREVA3SC5 Eleva- FIG. 2. Provisional map of Meserve Glacier, Wright Valley, Antarctica. tions and contours shown are probably all 38 m too high.

4 432 ISAGE wide, for 2 km down the valley wall, which slopes about 12, to the terminus at 440 m (Figs. 2 and 3). & %? mmm FIG. 3. Oblique aerial photograph of Meservc Glacier. Navy for U.S. Geological Survey. Photograph by U.S. The glacier terminates in a 20-m cliff which extends, with decreasing height, for 1200 and 2400 m along the east and west sides of the tongue, respectively. The surface of the tongue, along transverse lines, rises about 40 m from the cliff edge to the centreline. Gravity measurements indicate ice thicknesses of 40 to 60 m, with maximum values of 80 m, in each of four transverse profiles, the bed remaining at almost the same elevation. In one profile in the accumulation zone, the glacier is much thicker (350 m) on the eastern side than on the west, and it does occupy a significant depression. Meserve Glacier has been the site for glaciological investigations, including the flow law of cold ice (Holdsworth and Bull, 1970), and the mechanisms of formation of transverse crevasses (Holdsworth, 1969).

5 MASS BUDGETS : REGIONAL STUDIES 433 Investigations are being made (Everett and Behling, 1969) of the chemical and physical weathering characteristics of the prominent lateral moraines around the tongue in an attempt to obtain reliable indices of the relative ages of these and other glacial deposits in the ice-free areas of south Victoria Land. A necessary part of these studies is the assessment of the mass balance of Meserve Glacier. Field investigations were carried out during the and Antarctic field seasons. A detailed report on parts of the analysis is given by Carnein (1967). Measurement of mass balance parameters Climate of the area Temperatures. The mean annual air temperature at valley floor level (250 m) is about 16 C. In boreholes in Meserve Glacier at 480 m elevation the temperature below 8 m averages 18 C (Holdsworth and Bull, 1970). During the two summers, air temperatures were recorded at two sites, one, Moraine Station, on a flat part of the moraine 50 m east of the ice cliff, at about 450 m elevation, the other, Glacier Station, on the glacier surface, 90 m west of the moraine site, at 490 m elevation. About half of the temperature difference, 1 C, between these two stations is due to the elevation difference. The temperature data are summarized in Table 1. TABLE 1 MONTHLY MEAN AND EXTREME TEMPERATURES ( C), SUMMER , MESERVE GLACIER. THE FIRST NUMBER IN EACH PAIR REFERS TO MORAINE STATION; THE SECOND TO GLACIER STATION Mean monthly Mean maximum Mean minimum Absolute maximum Absolute minimum. Nov I (11-30 Nov.) Jan , , , , -fl , 12-2 Feb (1-13 Feb.) 61, , , , , 13-3 The minimum temperature during the 1966 winter at the Moraine Station was 40-5 C. During the summer, the temperatures in the ice-free valleys average about 0 C, the daily maxima occasionally reach 7 or 8 C, and a maximum of 12-3 C has been recorded (Bull, 1966). In these conditions, some melting occurs on the lower parts of glaciers, including Meserve Glacier and especially on the low-lying Wright Lower and Wilson Piedmont Glaciers, from which the Onyx River flows inland to Lake Vanda. Precipitation and winds. In Wright Valley the precipitation decreases westward: probably the annual precipitation does not exceed 5 g cm~ 2

6 434 ISAGE of water (all falling as snow) except on Wilson Piedmont Glacier. Almost all of the snowfall is associated with moist easterly winds from the McMurdo Sound area. During summer snow fell at Meserve Glacier on 8 out of 91 days, totalling about 1-2 g cm- 2. The direction of surface winds in Wright Valley is controlled to a large extent by the topography. At Meserve Glacier site summer winds are predominantly from the north-east. The frequencies are given in Table 2. TABLE 2 FREQUENCIES OF SURFACE WINDS AT MESERVE GLACIER FROM 6-HOURLY READINGS. THE FIRST NUMBER IN EACH PAIR REFERS TO MORAINE STATION (21 NOVEMBER FEBRUARY 1966); THE SECOND TO GLACIER STATION (25 NOVEMBER FEBRUARY 1966) Speed m sec" 1 N NE E SE S SW W NW Total % Total 0/ /o 1,4 6, 1 2,- ] 3 13,5 4,1 4,8 30,15 30,28 42,33 103, 84 6, 10 10,37 225,215 67,66 1,- 1,-, _ 2,0 1,0 6,7 12, 19 16,15 2, 1 1,3 1,3 38,48 11,15 4,2 3,3 9,8 2,- 2,- 20,13 6,4 6,8 13,16 10,8 2,2 -, 1 31,35 9, _ ,5 j _ -, 1 ; '_ 5,6 2,2 27,35 73,59 61, 54 47,36 111,90 6,10 13,40 338, 324 8,11 21,18 18,17 14,11 33,28 2,3 4,12 During the summer, the differences in frequencies of winds from different directions from month to month are not significant. The northeasterly winds usually decrease during the late evening; light southwesterly winds often blow between 0200 and At Lake Vanda, 15 km west of Meserve Glacier, easterly winds occur about two-thirds of the time, and westerlies for the remainder (Bull, 1966). The westerly winds are partly katabatic from the Antarctic ice sheet, warmed and reduced in relative humidity in their descent into Wright Valley. They are thin and frequently do not blow as far along the valley as Meserve Glacier. Strong westerly winds are probably much more common during the winter (Bull, 1966). Clouds and radiation. During the summer, low-level clouds, associated with the easterly and north-easterly winds, decrease in amount westward along Wright Valley. High-level clouds (cumulus and cirrus) usually come from the west and south-west. Cloud amounts at Meserve Glacier for summer are given in Table 3. Except for February, they are similar to those recorded at Lake Vanda in From the cloud amounts in , compared with those at the neighbouring Scott Base, where radiation measurements have been made, the radiation balance for the ice-free parts of Wright Valley has been estimated as a gain between 6000 and 29,000 cal cm- 2 yr" 1 (Bull, 1966). During the season measurements were made of the incoming

7 MASS BUDGETS : REGIONAL STUDIES 435 TABLE 3 MEAN CLOUD AMOUNTS (IN PER CENT) AT MORAINE STATION, MESERVE GLACIER, SUMMER Month 0700 Time Mean total clouds Estimated low cloud November 1-31 December 1-31 January 5-13 February C SO (i 3 C ß ,1 5 C S 20 a»,l 3 I Û I S 2 O 2 3 M s o 20 n x,i j " ' «> " ô ' '» " " '» MESEftVE GLACIER ANTARCTICA I - 6AS ST*TOJ I HUMCITY» Xi «63 I FIG. 4. Meteorological parameters at Base (Moraine) and Glacier Stations, Meserve Glacier, November 1965-February 1966.

8 436 ISAGE short-wave radiation. Outgoing long-wave radiation has been calculated by the methods of Ambach (1960), assuming values for the constants obtained by Andrews (1964) and Keeler (1964) at comparable latitudes in the northern hemisphere. The net flux of radiation on the glacier is plotted, with other weather parameters, in Fig. 4. Stake measurements of ablation and accumulation Twenty-five stakes in four lines across the ablation area, between 470 and 607 m elevation (Fig. 2), were measured weekly from late November 1965 to 13 February A denser line of stakes covered the 60 m closest to the eastern edge, over the tunnel excavated along the glacier base (Holdsworth and Bull, 1969); others were implanted in the cliff near the tunnel entrance. Most of the stakes were also measured during the summer. Twenty stakes in the accumulation area (Fig. 2) were measured on 4 and 17 December 1965 and 29 January Seventeen of them were remeasured on 29 November These stakes are not well distributed for accurate assessment of the mass balance, but the glacier shape is so simple that the necessary extrapolations should not introduce large errors. In the ablation zone, very little surface snow was present at the beginning of either field season. Most of the small winter snow accumulation is removed immediately by the strong winds. Table 4 gives the surface lowering at the stakes on the tongue. Those concentrated at the eastern edge are considered later. An ice density of 0-88 g cm" 3, measured at several surface sites, is assumed throughout. Table 4 shows that nearly 40 per cent of the annual loss occurred during the winter, from mid-february to mid-november. Ablation is largest close to the glacier edges: for the three stakes in each line nearest the glacier centreline, the summer lowering is 88 per cent of the mean at all stakes, and for the whole year is 90 per cent of the overall mean. The ablation rates on Meserve Glacier, about 1-4 g cm" 2 * per week during the summer, at air temperatures averaging 5 C, may be compared with others in Antarctica. On a blue-ice lobe of Reedy Glacier (86 S, 126 W), exposed to strong, dry winds, Mercer (1968,a) measured sublimation rates of about 0-7 g cm~ 2 per week with air temperatures averaging 20 C. Much lower rates are found on blue-ice fields in Dronning Maud Land (72 S, 3 W) (Schytt, 1961) and in the Ser Rondane mountains (72 S, 25 E) (Autenboer, 1964), despite much higher summer temperatures than in the Reedy Glacier area. Strong dry winds are much more important than "high" temperatures in producing sublimation at sub-zero temperatures. * All budget values given in water equivalent.

9 MASS BUDGETS : REGIONAL STUDIES 437 TABLE 4 SURFACE LOWERING OF STAKES DISTRIBUTED OVER THE GLACIER TONGUE Surface lowering (g cm" 2 ) Stake Elevation m from edge (E or W) m 27 Nov. 65 to 13 Feb Feb Nov. 65 to to 21 Nov Nov. 66 G E 143 E 242 E 211 W 118 W ; G E 40 E 139 E 238 E 224 W 106 W 55 W i G E 35 E 133 E 230 E 228 W 130 W 62 W I ; G E 108 E 198 E 196 W 99 W 33 W i Mean The observed variation of surface lowering with elevation has been extrapolated to cover the entire ablation zone. Average annual lowering near the terminus, at 460 m elevation, is about 34 g cm" 2, and is zero at the general balance line at 1130 m. The negative balance of the surface of the tongue totals x 10 6 kg. Attempts to measure the ablation (not including dry calving) on the ice cliffs have not been very successful. Only four markers on the eastern cliff were maintained for the entire 12-month period, November November The annual losses are given in Table 5, where they are compared with those on the upper surface nearby. The annual negative balance on the eastern cliff here is thus about 1-6 times that near the centreline. Assuming this same ratio holds all round the tongue, the annual loss by ablation from the cliff, not including dry calving, is about 9-5 X 10 6 kg.

10 438 ISAGE TABLE 5 ABLATION AT MARKERS ON AND NEAR THE EASTERN CLIFF, MESERVE GLACIER, 27 NOVEMBER NOVEMBER 1966 Marker Dl D2 CS1 CS2 Position 2 m above base of cliff 2 m above base of cliff 12 m above base of cliff 14 m above base of cliff Elevation in m ~460 ~ Surface lowering g cm~ I Mean 52-4 f J TS1 TS3 TS5 TS6 G4-0 On surface, 5 m from cliff edge On surface, 25 m from cliff edge On surface, 45 m from cliff edge On surface, 55 m from cliff edge On surface, 198 m from cliff edge In the névé basin during the period 4 December January 1966, surface lowering occurred at all stakes in the A' and B lines (elevations below 1350 m) except at A'l and B4. On the two higher traverses, C and D lines, summer accumulation occurred, except at the east side of the C line. Over the year to December 1966 positive balance was recorded at only 7 of the 16 stakes. Apart from B2 and C2, all stakes on the east side of the basin showed accumulation; the central stakes showed ablation; on the west side, four stakes showed negative and three showed positive balance (Table 6). About 1-2 km 2 of the glacier above the general balance line showed negative balance for the year; the remaining 6-9 km 2 showed a positive balance. The year , however, was probably not a "normal" year, the accumulation being much less than in the preceding years, as shown by the pit studies described below. Normally ablation exceeds accumulation only in the three small areas marked A in Fig. 2. The accumulation pattern in the névé basin is largely controlled by strong westerly winds in winter, which cause deflation in the western half and redeposition on the eastern half of the basin. The snowfall is so small that a single storm can cause large local areal variability of accumulation. For example, in one storm, on 6 December 1965, when southerly winds of more than 16 m sec" 1 were recorded at Moraine Station, 14-8 g cm" 2 was lost at stake C and more than 4-4 g cm" 2 at four other stakes on the eastern ends of B and C lines. Snow pit studies During the summer, snow stratigraphy studies were made in seven pits in the névé basin. Because the snowfall in the area is so small and so greatly reworked by strong winds, no reliable criteria could be developed for distinguishing annual units. In each of five of the pits (near stakes A, A', C, B and B2) a prominent, 12 cm-thick, fine-grained layer, probably wind slab, occurs at depths

11 MASS BUDGETS : REGIONAL STUDIES 439 between 80 and 120 cm. In these profiles loads above the slab are 35-0, 33-5, 41-0, 33-5, and 34-0 g cm- 2, so that if the slab were formed simultaneously at these five sites, the average accumulation since then has been areally uniform. Three of these sites showed loss during the year ranging from 0-2 to 9-5 g cm" 2 ; for the other two, data are not available. TABLE 6 CHANGES IN SNOW SURFACE LEVEL AT STAKES IN NÉVÉ BASIN, MESERVE GLACIER, Stake A'l A'2 A' A'3 A'4 Elevation* m December January 1966 g cm" December December 1966 g cm" 2 -f Bl B2 B B3 B Cl C2 C C3 C Dl D2 D D * Elevations are approximate, pending corrections to preliminary survey calculations. In some of the profiles alternations of density are discernible. If successive dense layers mark annual layers, the accumulation at B in some recent year was 17 g orr 2. It is noteworthy that in pits dug in areas showing surface lowering in (B2 and C), no ice or high density snow layer was found to depths of 2 m. Attempts to determine the annual accumulation from the increase with depth of the grain size, using data from Little America (Vickers, 1966) and the McMurdo-South Pole traverse (Giovinetto, 1963) suggest that in the accumulation zone the positive balance normally averages between 3 and 7 g cm" 2 yr" 1. For the year , an average gain of 3-0 g cm- 2 is estimated for the 6-9 km 2 area of the névé basin showing positive balance. In the 1-2 km 2 showing negative balance the average loss was about 3-1 g cm" 2.

12 440 ISAGE Dry calving Dry calving from the cliff occurs all around the ablation tongue, forming an apron of broken ice, averaging 2 m thick and extending 12m from the cliff. The total mass of ice in this apron is about 30-0 x 10 6 kg, almost all avalanche debris. For the 12-month period January 1966-January 1967, the total natural calving along the western and eastern cliffs were 1-06 x 10 G kg, and 2-11 x 10 6 kg, respectively. In November 1966, 0-47 X 10 6 kg was blasted from the eastern cliff, above the tunnel entrance. As this material would probably have calved naturally during the summer, it is included in the annual mass lost by calving, 3-64 x 10 6 kg. TABLE 7 MASS BALANCE OF MESERVE GLACIER, NOVEMBER 1965-NOVEMBER 1966 I Mass x 10" kg Negative balance of surface below general balance line ; Negative balance of cliff, excluding dry calving : 9-5 Dry calving from cliff : 3-6 Negative balance in névé basin (1-2 km 2 ) Total losses Positive balance in névé basin (6-9 km 2 ) For whole glacier, loss This is equivalent to an average loss of 0-61 g cm" 1 over the whole glacier area. Estimate of mass balance For the year the quantities in the mass balance equation are given in Table 7. In "normal" years the positive balance in the névé basin is probably between 3 and 7 g cm- 2. Assuming that the loss on the tongue, 230 X 10 6 kg, also applies to that "normal" year, the glacier would have a positive balance between 13 and 237 x 10 6 kg, with the smaller figure being the more likely. As the total mass of the glacier is about 1-5 X kg, none of these estimated values of the mass balance represents a significant departure from equilibrium. Consideration of some of the mechanisms of ablation Melting On clear summer days, even when air temperatures are as low as 7 C, some melting does occur on the glacier surface, the ice cliff and the apron, forming meltwater streams on both sides of the glacier. The total runoff in the stream near the tunnel entrance, for the period 14 November February 1966, was about 60,000 m 3 (Carnein, 1967). Most of this flow was due to melting of the relatively dirty avalanche debris apron. On one day the total runoff over the eastern cliff, measured directly, was 60 to 100 m 3, only 4 per cent of the runoff in

13 MASS BUDGETS." REGIONAL STUDIES 441 the stream near the tunnel entrance. The total whole summer runoff from the east side of the glacier surface was about 2400 m 3, giving an average loss of about 4 g cm- 2 over the ice cliff and the 100 m-wide strip at the edge of the glacier, below about 700 m elevation, where melting was observed. Meltwater thus constitutes only 2 or 3 per cent of the mass lost from the glacier tongue. Evaporation and sublimation At Glacier Station measurements were made during the summer of air temperature, relative humidity and wind velocity and direction at one height, 170 cm. The evaporation that would occur, assuming an icesurface temperature = 0 C, vapour pressure = 6-11 mbar and wind velocity = 0, has been estimated, using the formulae of Lister and Taylor (1961) and Keeler (1964). For three representative two-day periods in early, middle and late summer, evaporations were calculated from mean hourly values of the parameters; for the rest of the summer, evaporation rates have so far been calculated only from mean daily values. The calculated evaporation for the summer is 8-36 g cm" 2, 50 per cent of the measured loss of 16-4 g cm" 2 at five stakes about 5 m from the meteorological station. Here the meltwater runoff is about 3-3 g cm" 2, 20 per cent of the total summer loss. The remaining 30 per cent is unexplained at present. From 13 February to 21 November 1966 the loss at these five stakes averaged 12-8 g cm" 2. A very small part of this loss may have been by evaporation after 13 February, but most of it is due to sublimation caused by the strong, dry, westerly winds. The stability of the ice cliff Along 450 m of the eastern margin between the G4 and Gl lines, the ice flow over the line of the cliff has been calculated, assuming that the variation of velocity with depth is everywhere of the same form as in borehole Ml, close to stake Gl (Holdsworth and Bull, 1969), and that the velocity at depth is in the same direction as at the surface. The balance for the 450 m-long cliff section is as follows: Throughflow of ice past line of cliff = 2-84 x 10 6 kg Loss of ice by dry-calving = 0-97 x 10 6 kg Loss of ice by other ablation =2-14 x 10 6 kg Net loss = 0-27 x 10 6 kg This loss, if real, would produce an annual cliff retreat of about 4-9 cm, comparable with the errors in the various estimates. It can be said only that the cliff is approximately in equilibrium. The stability of the glacier surface The equilibrium state of the glacier may also be assessed from the mass balance of a section of the ablation zone. The annual flow of ice through.

14 442 1SAGE the glacier sections along the G2 and G4 lines has been calculated from the components of the surface velocities perpendicular to the lines, multiplied by the ice thickness and a factor between 0-83 and This factor is obtained by extrapolating over a small range of depths the ratio of the throughflow to the surface velocity times depth, in holes Ml and M2 (Fig. 2), where the ice thicknesses are 22-9 and 34-8 m, and the values of this factor are 0-83 and 0-86, respectively. The throughflow (cm 3 cm" 1 yr" 1 ) at each of the movement stakes was plotted against position, and the total throughflow for the two sections was calculated by a graphical method. Similarly the surface lowerings at the stakes in the G2, G3 and G4 lines were used to estimate the mass loss from the glacier between the G2 and G4 lines. The values obtained for the one-year period are as follows: Ice flowing through G2 section = X 10 9 cm3 Ice flowing through G4 section = 54-4 X 10 9 cm3 Difference = 48-6 X 10 9 cm 3 s = 43-8 x 10«kg Negative balance, top surface = 36-0 X 10 6 kg Dry calving, cliff = 0-8 X 10 6 kg Other ablation, cliff = 3-0 X 10 6 kg Total losses = 39-8 X 10 6 kg Overall gain = 4-0 X 10 6 kg yr-i Over the 14-6 x 10 8 cm 2 area, this gain is equivalent to an annual increase in surface elevation of 3-1 cm, about the same as errors in the estimates of surface lowering and cliff loss. Again it may be concluded only that this part of the glacier is approximately in equilibrium. Discussion The examination of the mass balance for the entire glacier, the stability of the cliffs and of the surface elevation of the tongue, all indicate that the Meserve Glacier is very nearly in equilibrium, at least for this very limited period of observation. With such a glacier, on which the accumulation is so low, however, small changes in the accumulation or ablation can produce very large changes in the glacier. An increase of only 3 g cm" 2 yr -1 in the mean accumulation in the névé basin, with unchanged ablation, would, if maintained for several hundred years, extend the tongue about 0-7 km downslope and laterally to a position between the innermost and middle prominent lateral moraines shown in Fig. 3. Similarly, a reduction in the accumulation by 2 g cm" 2 yr" 1 could cause the tongue to retreat 1 km, leaving only a short tongue, as now occurs in the unnamed glacier, 4 km west of Meserve Glacier. Farther west, the glaciers, all originating in basins similar to that of Meserve Glacier, become progressively smaller, and near the head of the Wright Valley, 25 to 40 km west of Meserve Glacier, the cirques are almost empty (Bull et ai, 1962). Measurements of annual accumulation in other névé basins along Asgard Range would be very valuable. At present the only

15 MASS BUDGETS : REGIONAL STUDIES 443 measurements from the area are those on Meserve Glacier, values of & and 6 g cm" 2 yr -1 on Wilson Piedmont Glacier, within 12 km of the coast (Bull, 1966), and 5 to 7 g cm- 2 at 800 to 1000 m elevation on Packard Glacier, Victoria Valley, 30 km from the coast (Calkin, 1963; reported; in Bull, 1966). On the névé basins of Commonwealth and Canada Glaciers, which flow south from Asgard Range into Taylor Valley, 12 and 20 km from the coast respectively, Henderson et al. (1966) estimate mean positive balances of about 2-8 nd 4-6 g cm~ 2 yr" 1, assuming that the glaciers are in equilibrium. The near-equilibrium state of Meserve Glacier does not permit any profitable calculation of the rate of lateral recession of the tongue, or of the age of the innermost of the lateral moraines, on which Everett and Behling (1969) have been studying chemical and physical weathering. In these moraines, in the ice-cemented debris under Meserve Glacier, and in most of the surface deposits in the ice-free regions, are accumulations of salts, many of them soluble. The possible origin of these salts include trapped sea water, hot-spring activity associated with the volcanism, leaching of evaporite beds in sedimentary rocks, wind-transported marine salts, and chemical weathering of local bedrock and soil. Jones and Faure (1967) find that although the Sr 87 :Sr 88 ratio in surface snow of Meserve Glacier is nearly identical with that of Ross Sea water, the ratio for water from Onyx River and Lake Vanda and for valley soils is similar to that of the major bedrock units. They conclude that as no local evaporite beds have been found, the strontium in Lake Vanda has been derived largely by chemical weathering of bedrock. The time of formation of the salts by chemical weathering is still uncertain. Everett and Behling (1969) argue that the salts in the moraines around Meserve Glacier are being formed, in situ, by chemical and physical weathering. Although meltwater does leach some of the salts, they believe that the salt concentrations can be used at least to indicate relative age. On the other hand, Faure (oral communication) considers it more likely that the salts are a relict feature of a former warmer and more humid environment, perhaps in the Tertiary. Part of his argument is based on the high salt concentrations in the till and in the basal ice exposed in the tunnel under Meserve Glacier (Holdsworth and Bull, 1969). However, the glacier could advance 400 m (the distance from the tunnel to the snout) very quickly, by a slight change in the annual accumulation. The process of advance, by increasing amounts of dry-calving until the calved material becomes incorporated into the moving glacier, need not destroy or rework the debris and salts exposed on the surface beyond the end of the former glacier. On the other hand, the ice-free areas of south Victoria Land were formerly wetter (and therefore probably warmer). For example, Calkin (1964) in Victoria Valley and Bull Pass found many inactive mud-flows, and solifluction lobes much larger than the few now active. One of these; large flows overrode algae that give a radiocarbon age of 9700 years.

16 444 ISAGE Everett and Behling's conclusions are supported by Jones (unpublished studies). High on the west side of Meserve Glacier the morainal material is mainly basalt. The Sr 87 : Sr 86 ratio of the water-soluble salts in the soils here is about 0-707, indicating the salts are derived largely from the basalts. At lower elevations, where the moraine contains more debris of metasedimentary and granitic rocks, the Sr 87 :Sr 86 ratio in the salts increases, reflecting the composition of the moraine, and suggesting that the salts have formed since the moraines were deposited. At present, however, no reliable method exists of dating the moraines. It is difficult to explain the size of the lateral moraines of Meserve Glacier: the basal ice layers contain very little debris (Holdsworth and Bull, 1969), and, the ice at the contact is not moving, so that little erosion is now taking place. No significant morainal deposits are found immediately adjacent to the cliff although there are small glacier-ice cored "incipient moraines" adjacent to the glacier at the lower end of the accumulation zone. The debris is probably being carried to the glacier edge along discrete planes in the basal ice (Behling, oral communication). For significant amounts of erosion to occur, the temperature at the glacier bed must be close to the melting point, which, due to the salt content, may be below 0 C. For this to occur, however, either the glacier thickness must increase by more than 600 m (assuming an average value for geothermal heat flow), or the mean annual air temperature must increase by about 10 C, or by a combination of these factors. When the outermost moraines formed, the glacier tongue could have been three times as thick as now, an increase of 200 m. Mercer (1968a) has evidence, from lake sediments and solifluction near Reedy Glacier (86 S, 130 W) of an environment 6 to 10 C warmer than at present. In those conditions the Ross Ice Shelf and probably most of the West Antarctic Ice Sheet would disappear (Mercer, 1968b). Mercer considers that an isotopically-dated high sea level about 120,000 years BP, probably during the Sangamon interglacial, dates the most recent time in which this increased temperature is likely to have occurred. One of the prominent lateral moraines of Meserve Glacier may date from the same time. On the other hand, Dort (1969) suggests that the Hypsithermal period (about 6000 years BP) of higher temperatures was the time of the most recent advance, and that the cooling that caused the Little Ice Age in temperate regions produced an increased amount of ice in the Ross Sea area, a reduction in the snowfall, and a retreat of the alpine glaciers in southern Victoria Land. It is quite apparent that the prime need in studies of the glacial history of the Wright Valley and the glaciology of the alpine and outlet glaciers is a reliable method of age determination. Everett and Behling's studies of weathering processes hold promise of a good method of relative age determination but absolute ages remain a problem.

17 MASS BUDGETS : REGIONAL STUDIES 445 Acknowledgements All the members of the and parties assisted with the data collection. Thanks are due particularly to Gerald Holdsworth, Principal Investigator of the project, for his painstaking guidance of the field work, and for his careful estimates of the dry-calving and meltwater runoff. The research on Merserve Glacier has been supported by Grants GA-205 and GA-532 awarded to The Ohio State University Research Foundation by the National Science Foundation. The assistance of the personnel of the U.S. Antarctic Research Program and the U.S. Navy is very gratefully acknowledged. Dr. John Mercer, Robert Behling, and Dr. Lois Jones have critically reviewed parts of the manuscript. REFERENCES AMBACH, W Investigations of the heat balance in the area of ablation on the Greenland Ice Cap. Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B, Bd. 10, p ANDREWS, R. H Meteorology and heat balance of the ablation area, White Glacier. Axel Heiberg Island Research Reports, Meteorology, No. 1, Montreal, McGill University. AUTENBOER, T. VAN The geomorphology and glacial geology of the Sor Rondane, Dronning Maud Land. In: Antarctic Geology: First International Symposium on Antarctic Geology, Proceedings, Capetown. Adie, R. }., Ed. 1963, New York, John Wiley and Sons, Inc., p BULL, C Climatological observations in ice-free areas of southern Victoria Land, Antarctica. In: Studies in Antarctic Meteorology. Rubin, M. J., Ed., Washington, D.C. American Geophysical Union, p (Antarctic Research Series, Vol. 9). BULL, C, MCKELVEV, B. C. and WEBB, P. N Quaternary glaciations in southern Victoria Land, Antarctica. /. Glaciol., Vol. 4, No. 31, p CALKIN, P. E Geomorphology and glacial geology of the Victoria Valley system, southern Victoria Land, Antarctica. (Ph.D. dissertation, The Ohio State University). CALKIN, P. E Geomorphology and glacial geology of the Victoria Valley system, southern Victoria Land, Antarctica. Institute of Polar Studies Report No. 10, Columbus, The Ohio State University. CARNEIN, C. R Mass balance of the Mescrve Glacier, Wright Valley, Antarctica. (M.S. thesis, The Ohio State University). DORT, W Climatic causes of alpine glacier fluctuations, southern Victoria Land. In: Gow, A. J. et al., Eds., International Symposium on Antarctic Glaciological Exploration (ISAGE), Hanover, New Hampshire, U.S.A., 3-7 September Cambridge (Pub. No. 86 of IASH), p EVERETT, K. R. and BEHLINO, R. E. Chemical and physical characteristics of the Meserve Glacier morainal soils, Wright Valley, Antarctica: an index of relative age? In: Gow, A. J. et al., Eds., International Symposium on Antarctic Glaciological Exploration (ISAGE), Hanover, New Hampshire, U.S.A., 3-7 September Cambridge (Pub. No. 86 of IASH), p GIOVINETTO, M. B Glaciological studies on the McMurdo-South Pole Traverse 1960^61. Institute of Polar Studies Report No. 7, Columbia, The Ohio State University. HENDERSON, R. A., PREBBLE, W. M., HOARE, R. A., POPPLEWELL, K. B., HOUSE, D. A. and WILSON, A. T An ablation rate for Lake Fryxell, Victoria Land, Antarctica. J. Glaciol., Vol. 6, No. 43, p HOLDSWORTH, G Primary transverse crevasses. J. Glaciol., Vol. 8, No. 52, p

18 446 ISAGE HOLDSWORTH, G. and BULL, C The flow law of cold ice; investigations on Meserve Glacier, Antarctica. In: Gow, A. J. el al., Eds., International Symposium on Antarctic Glaciological Exploration USAGE), Hanover, New Hampshire, U.S.A., 3-7 September Cambridge (Pub. No. 86 of IASH), p JONES, L. M. and FAURE, G Origin-of the salts in Lake Vanda, Wright Valley, southern Victoria Land, Antarctica. Earth and Planetary Science Letters, Vol. 3, p KEELER, C. M Relationship between climate, ablation and runoff on the Sverdrup Glacier, 1963, Devon Island, N.W.T. Arctic Institute of North America, Research Paper No. 27. LISTER, H. and TAYLOR, P. F Heat balance and ablation on an Arctic glacier. Meddelelser om Granland, Bd. 158, Nr. 7. MERCER, J. H. 1968a. Glacial geology of the Reedy Glacier area, Antarctica. Geological Society of America Bulletin, Vol. 79, p MERCER, J. H. 1968b. Antarctic ice and Sangamon Sea level. In: WARD, W., Ed., International Union of Geophysics, General Assembly of Bern, 25 Sept.- 7 Oct Pub. No. 79 of IASH, p SCHYTT, V Blue ice-fields, moraine features and glacier fluctuations. Part E of Glaciology II, Norwegian-British-Swedish Antarctic Expedition, , Scientific Results, Vol. 4, p Oslo, Norsk Polarinstitutt. VICKERS, W. W A study of ice accumulation and tropospheric circulation in Western Antarctica. In: Studies in Antarctic Meteorology, Rubin, M. J., ed., Washington, D.C. American Geophysical Union, p (Antarctic Research Series, Vol. 9).