IMMEDIATE SCIENCE REPORT
K047: Climate and Landscape History from shallow Drilling in the Dry Valleys
Antarctica New Zealand
1 Popular Summary of Scientific Work Achieved
A model to explain the occurrence of ground ice in glacial sediments and bedrock at high altitudes (>1000m) throughout the Dry Valleys where liquid water is rare was developed from work on Table Mtn. (Dickinson and Grapes 1997; Dickinson and Rosen 2002). The model is based on mineralogical, chemical and isotopic analyses of ground ice and frozen sediments that come from cores of Sirius Group sediments at Table Mtn. It indicates that the ground ice and diagenetic minerals accumulated over long periods of time from atmospheric water vapour and brine films formed on the surface of the ground. Although this model may apply at Table Mtn. for the very old glacial sediments of the Sirius Group, it has yet to be tested at other locations in the Dry Valleys.
The sampling programme of the 2001/02 season aims to test the Table Mtn. model by examining soils and ice cemented sediments from three geologically different locations, which are in close proximity to each other in the Dry Valleys. The three areas included: Beacon Valley for its polygonal ground, glacial sediments and old ice, Arena Valley for its potentially old, non-glacial soils, and Pearse Valley for its abundance of young glacial sediments at a low elevation. Evaluation of analytical results may lead to shallow core drilling of certain sites in the future to test the Table Mtn model.
2 Proposed Programme
To understand the processes by which ground ice is formed, a comparative set of samples is needed from a variety of ground surfaces. These surfaces must lie on a transect from low to high elevation and extend through a range of ages. In sampling and analyzing the soils and their underlying ice cemented sediments there were three aims: 1) To determine if a chemical and mineralogical relationship exists between the soils and ice cemented sediments. 2) To determine if there are differences in the chemistry and mineralogy of the soils and ice cemented sediments between the different areas. 3) To determine the relationship of relative soil age, chemistry and ice content to polygonal ground. Evaluation of analytical results may lead to shallow core drilling of certain sites in the future.
3 Scientific Endeavours and Achievements
The three event people spent 21 days in the field from 15 Nov to 5 Dec. Dickinson spent an additional day (6 Dec) with Alex Pyne (VUW) and Paul Houston (Antarctica NZ) at Table Mtn. collecting samples and re-programming the data loggers on two temperature probes for 2002. A total of 21 soil pits were described and sampled in Beacon, Arena, and Pearse valleys (Table 1). Soil pits were generally dug in the centers of polygons to control the comparison between different sites and areas. Polygon centres are thought to be the least active area and hence should contain the oldest most chemically developed soil.
The following method was used in digging most all of the soil pits: In the area to be excavated (1m × 1m × .5m), the surface material was scrapped off and placed on a 2m square polythene tarpaulin. The underlying soil was then dug out and placed on another 2m square polythene tarpaulin. Loose soil material was dug to a depth of one metre or the top of the ice-cemented soil which ever was the shallowest. The top of the ice-cemented material was sampled by using a gasoline powered hammer drill to excavate fist-sized samples. After the soil profile and permafrost were described and sampled, all material from the respective polythene tarpaulins was replaced. The ground surface was raked and swept to restore as much as possible of the original appearance. Analyses of the samples will include; thinsections of soil clods as well as major cation and anion chemistry of soluble salts in the soils and ice from the permafrost. These methods were used to achieve the three aims listed above.
The term "ground ice" refers to all types of ice formed in freezing and frozen ground (Permafrost Subcommittee, 1988 p 46). Permafrost refers to the permanently frozen (<0° C) condition and includes both dry and wet (ice) materials. Subsurface conditions in the Dry Valleys are generally different from those in arctic and alpine environments in that there is usually 30 to 60 cm of dry frozen sediments above ice cemented sediments. However, because most workers think of permafrost as ice cemented, the term permafrost in this report will include only the ice cemented materials.
Beacon Valley together with its 12 side valleys (McKelvey and Webb 1959) have the most extensive and best defined polygonal ground in the Dry Valley area. Elevations of the main valley floor are between 1300 and 1500 m while elevations of the side valleys are about 200 m higher. Winds during the field visit were generally down valley and less than 10 knots, however, diurnal up-valley winds were also encountered. Our camp location was on a small patch of snow on the southwestern flank of University Valley (1650m; S77°51.368′ E160°41.987′) and was selected for the snow patch and central location to the valley. However, due to the rugged terrain and subsequent slow walking, it probably would have been more convenient and as climatically comfortable to camp adjacent to the main valley bottom on the southeastern flank.
The polygonal ground of Beacon Valley was studied in the 1960′s (Berg and Black 1966) and soils of the area have been described by (Bockheim 1982; Bockheim and Ugolini 1972; Linkletter et al. 1973; Potter and Wilson 1983; Ugolini et al. 1973). More recently, weather stations with ground temperature probes have been installed in the lower and central parts of the main Beacon Valley along with strain gauges across polygon troughs (B. Hallett and R. Sletten, pers. comm). In addition rock glaciers, which are uncommon in the Dry Valleys, emerge out of Friedmann and Mullins valleys onto the floor of Beacon Valley.
The origin of the debris material on the floors of Beacon Valley and its side valleys remains unclear. Although it appears that at one time a tongue of the Taylor Glacier must have occupied the valley floor, there is page 4 no obvious moraine to support this supposition. In addition, recent drilling and ground penetrating radar (GPR) indicate that debris-laden ice lies below 2 – 3 m of ice cemented debris on the floor of Beacon Valley. The thickness of this ice is unknown because surface salts obscure GPR results (R. Sletten, pers. comm.) but it may be over 150 m thick (A Hubbard, pers. comm.). (Marchant et al. 1996; Sugden et al. 1995) dated volcanic ashes associated with this ice and have suggested that it is more than 8 Ma old.
Polygons on the floor of Beacon Valley have a 10-20 m diameter and 2 – 3 m height differential between trough and polygon centres. Although the diameter of the polygons is not uncommon, the large height differential is and may result from a long development period or the glacial ice core of the valley. The height differential of the polygons in the side valleys is less and does not appear to exceed 1 m. Adjacent to and along the southeast flank of the main valley floor, is a lateral strip that is either absent of polygons or has polygons with the least amount of relief in the area. Depth to ice cemented ground appears to be 40 – 60 cm throughout the area. In one of the rock glaciers, clear ice was found below the ice cemented ground at about 30 cm.
The activity of any single polygon or part of it may be reflected by the distribution of the material in the troughs. Parts of troughs are flat having been filled with sand while other parts are steep and rocky with angular cobbles and boulders. This angular material may be sorted or unsorted. On the active part of a polygon, clasts may roll off the steep sides and into the trough. Sorting of clasts in the trough may occur by what the center crack is able to accommodate. On the inactive part of a polygon, wind blown sand may accumulate in the trough. This observation suggests that polygon activity may be dynamic so parts of it are active while at the same time other parts are inactive.
A major problem apparent from the field work is to understand what controls the age of the surface and the relationship to polygon development in Beacon Valley. Alternatively, it may be the ice content below the surface that controls polygon development. Soil development and age may be more of a function of the material, aspect and moisture regime, rather than the depositional age of the material in which the soil is forming. The absence of recognizable glacial deposits in Beacon Valley may reflect the activity of the polygons which has destroyed the structure of the moraines making them unrecognizable.
In contrast to Beacon Valley, Arena Valley has little polygonal ground and is mostly underlain by a bedrock of Ferrar Dolerite and Beacon Supergroup sediments. Elevations of the main valley floor are about 200 m lower that Beacon Valley and lie between 1100 and 1300 m. Winds, generally down valley, were 10 – 15 knots stronger than Beacon Valley during the field visit and probably averaged between 15 and 25 knots. Our camp was located at the western end (generally the leeward end) of a linear snow patch at the northern edge of Ashtray Basin (1130m; S77°51.593′ E160°56.915′) and was selected for the snow patch and central location to the valley. Wind strength and duration at this location was about average for the valley floor.page 5
Most soils in Arena Valley have developed directly on bedrock but in some areas they have developed on talus and scree which have slid off steep valley sides. Except for the terminal moraines marking the retreat of the Taylor Glacier at the mouth of the valley, there are no glacial deposits in the valley. Polygonal ground probably covers less than 10% of the area of the valley bottom and slopes, and appears to be restricted to talus and scree deposits. This would suggest that the lack of polygonal ground in Arena Valley is due mainly to the lack of loose material upon which it can develop. However, local climatic conditions cannot be excluded because most of the polygonal ground also appears to be in areas of increased moisture and snow accumulation. In addition, the overall windy nature of the valley may remove much of the moisture making polygonal ground more difficult to develop.
In addition to talus and scree, polygonal ground was also found to form on sand dunes climbing up the sides of the valley (sites AV-4-6). Pits in the terminal moraines of the Taylor Glacier, showed muds and silts with horizontal bedding that suggests small lakes or ponds originated from glacial meltwater.
Soils in Arena Valley should be relatively older than those in Beacon Valley because of the mobility of polygonal ground that would homogenize soils rather than promote horizonation. Therefore, soils in Arena should be more horizonated both chemically and physically than those in Beacon Valley. Most pavement surfaces in Arena are similar to that apolygonal ground which is found on the southeastern flank of Beacon Valley with well-sorted pebble sized and highly ventifacted pavement surface. The main question in Arena Valley is why there is such an absence of glacial deposits when the adjacent Beacon Valley has been so highly glaciated. In addition, why is polygonal ground absent from the few glacial deposits that are present in Arena Valley.
In contrast to Beacon and Arena valleys, the elevation Pearse Valley is much lower and ranges between 400 and 500 m MSL. Our camp was on an alluvial terrace located at the eastern edge of Lake House (325m; S77°42.101′ E161°26.924′) and was selected for its proximity to a source of water. Wind direction and strength seems highly variable throughout the valley and diurnal variations were common. During the field visit, winds did not exceed 20 knts and seemed strongest from 2 – 5 am. In general wind strength and duration were in-between those of Beacon and Arena valleys.
Pearse Valley contains mostly glacial deposits representing the retreat of the main Taylor Glacier and subsequent retreat of the lateral valley glaciers. As a result, the chemical development of the soils should reflect this Holocene deposition and contrast to the older soils in Beacon and Arena valleys. Polygonal ground covers 40 – 50% of the valley floor and slopes, however, this was difficult to estimate due to the lack of snow in polygon troughs.
About 10% of the valley floor is covered by sand from eolian deposition and this does not include numerous pockets of sand lodged in troughs of polygons and in other sheltered areas. Much of this sand is protected by a lag of 5 – 8 mm grannules and therefore is not mobile under winds of about 50 knts. Much of the sand probably came from stream page 6 systems draining meltwater from the retreating glaciers. The main sand dune, climbing the northeast slope of the valley apparently has brine flowing on top of ice cemented sand which accumulates in salt pond (dry on the surface) at the base of the dunes. Ice cemented samples from the dune and brine from approximately 50 cm deep in the salt pond were taken for chemical analysis.
Depth to ice cement and clear ice under moraines varies in the valley from 0.25 m to >1m and was encountered in every pit except PV-LAK1, the sediments of which may represent and old lake deposit. It is not clear what factors control the depth to ice cement but aspect and moisture regime do not seem to have a direct relationship. In addition, the degree of polygonal ground development does not appear to be related to the depth of the ice cement. For example, two pits were dug in the vicinity of PV-7, one in well developed polygonal ground and the other in poorly developed polygonal ground and ice cement was found at 25 cm in both pits.
Perhaps our most interesting find in Pearse Valley is the presence of clear ice in pits PV-1, 2, 3 & 6. This ice possibly represents and ice cored moraine which may have derived from the Schlatter Glacier. The surface of this ice is smooth and it is not clear how the contact between it and loose sand above can be so sharp. Why there is not ice cemented sand above, suggests the clear ice is ablating under the sand. Although the clear ice seems to have a limited extent, it may have a greater extent if it lies below ice cemented soil in other parts of the valley.
A preliminary report on the soil profile descriptions and profiling will be published as an Antarctic Research Centre Report in June 2002. This report will include much of the technical work on the drilling system, core logs and photographs, maps and cross sections. Copies of this report will be sent to Antarctica NZ.
Further publications of the scientific results will be published in international peer-reviewed scientific journals. Copies of this work will also be sent, when available, to Antarctica NZ.
Thanks to the following: Prof Peter Barrett, (Director, Antarctic Research Centre, VUW) Dean Peterson, Paul Woodgate and Jim Cowie, (Antarctica NZ) All of the Scott Base personnel (Nov 2001 - Jan 2002) Bain Webster and Jeff Ashby (Webster Drilling Inc, NZ)
Funding and Support Antarctica New Zealand, Strategic Development Fund, VUW Webster Drilling Inc, NZ
Berg, T.E., and Black, R.F., 1966, Preliminary measurements of growth of non-sorted polygons, Victoria land, Antarctica, in Tedrow, J.C.F., ed., Antarctic Soils and Soil forming Processes: Antarctic Research Series:, American Geophysical Union, p. 61-108.
Bockheim, J.G., 1982, Properties of a chronosequence of ultraxerous soils in the Trans-Antarctic mountains: Geoderma, v. 28, p. 239-255.
Bockheim, J.G., and Ugolini, F.C., 1972, Chronosequences of soils in the Beacon Valley, Antarctica, in Adams, W.P., and Helleiner, F.M., eds., International Geography:, p. 301-303.
Dickinson, W.W., and Grapes, R.H., 1997, Authigenic chabazite and implications for weathering in Sirius Group diamictite, Table Mountain, Dry Valleys, Antarctica: Journal Sedimentary Research, v. 67, p. 815-820.
Dickinson, W.W., and Rosen, M.R., 2002, Antarctic ground ice and diagenetic minerals from atmospheric moisture and brine films: Nature, v. (in review).
Linkletter, G.O., Bockheim, J.G., and Ugolini, F.C., 1973, Soils and glacial deposits in the Beacon Valley, southern Victoria Land, Antarctica: New Zealand Journal of Geology and Geophysics, v. 16, p. 90-108.
Marchant, D.R., Denton, G.H., Swisher, C.C.I., and Potter, N.J., 1996, Late Cenozoic Antarctic paleoclimate reconstructed from volcanic ashes in the dry valleys region of southern Victoria Land: Geological Society of America Bulletin, v. 108, p. 181-194.
McKelvey, B.C., and Webb, P.N., 1959, Geological investigations in southern Victoria land, Antarctica. Part 2 - Geology of the upper Taylor Glacier region: New Zealand Journal of Geology and Geophysics, v. 2, p. 718-728.
Potter, N.J., and Wilson, S.C., 1983, Glacial geology and soils in Beacon Valley: Antarctic Journal of the US; 1983 Review, v. 18, p. 100-103.
Sugden, D.E., Marchant, D.R., N., P., Souchez, R.A., Denton, G.H., Swisher, C.C.I., and Tison, J.-L., 1995, Preservation of Miocene glacier ice in East Antarctica: Nature, v. 376, p. 412-414.
Ugolini, F.C., Bockheim, J.G., and Anderson, D.M., 1973, Soil development and patterned ground evolution in Beacon Valley, Antarctica: Permafrost: North American Contribution to the Second International Conference, p. 246-254.