| Summary: | A variety of Antarctic lakes occur either as proglacial lakes to glaciers, or as detached lakes within an enclosed drainage system in the Dry Valleys of McMurdo Sound, Antarctica. Some of these are the evaporated remnants of much larger proglacial lakes that formed during the last glacial, when the Ross Ice Shelf thickened, grounding on the continental shelf, and advanced landward entering many of the Dry Valleys. These former proglacial lakes have been the source of datable material using U/Th methods on the products of evaporation (CaC0₃), and ¹⁴C on algal deposits associated with many lacustrine sediments. These methods provide dates for glacial advances and hence allow correlation with similar northern hemisphere events. Interpretation of both the data generated, and morphologic evidence for paleo-lacustrine environments, requires a knowledge of how these former lakes developed and the processes occurring within them. To this end a modem proglacial lake, Trough Lake in Pyramid Trough, was selected to study and provide a model for sedimentation in the former lakes as well as a guide for interpretation of former proglacial environments. Although Trough Lake is proglacial to an alpine glacier, not a large ice sheet, and it is much smaller than the former lakes, it makes a satisfactory surrogatte. The principle aim was to examine the sedimentary processes occurring within the lake and to use this information to construct a facies model for the interpretation of former pro glacial lacustrine environments. The transport of ice-rafted, formerly glacial material, across the lake ice and the changes that occur to it have been examined. The results indicate that aeolian processes, as well as extraction of the fine fraction through the lake ice, occur concentrating the coarse fraction in ice-cored domes at the distal end of the lake. The fine fraction if it migrates through the lake ice becomes part of the normal, predominantly silt, lacustrine sediments. However, the distance a sand particle can move through the ice is size dependent, the coarser the material the smaller the distance it can migrate downward. This sand/ice interaction manifests itself in bimodal lacustrine sediments becoming progressively coarser away from the glacier until a point is reached where all the sand fraction that can migrate through the ice, has done so, and beyond this, lacustrine sediments will be unimodal. Thus it is apparent, that grain size distribution analysis of proglacial lacustrine sediments, and their associated drift, can be very useful in determining the size and structure of former Antarctic proglacial lakes. The calving of icebergs from the glacier snout and their eventual replacement by lake ice provides a mechanism by which lake ice can move at a rate of about 18 m per year (at Trough Lake) away from the glacier. The formation of the characteristic ice-cored domes found on many pro glacial or former proglacial lakes in the Mcurdo Sound area was demonstrated in a laboratory experiment designed to model the Antarctic environment. These domes are important as they often cover and preserve underlying lacustrine silts and algae when left behind by falling lake levels through retreat of the associated glacier. In addition these deposits can be used to gauge the extent of former lakes as well as providing samples for ¹⁴C dating. Evidence of seasonal cracks within the lake was observed and a core of sediment deposited through such a crack was extracted. This suggests another method by which lacustrine silts may be preserved in a very dry and windy Antarctic environment The physics of sand/ice interactions was examined in some detail, and it was demonstrated that the feasibility of sand moving through ice was not size related as had been suggested by some workers, however the rate at which sand moves through ice is inversely proportional to grain size. This provides a non-structural mechanism by which sand can move through lake ice and accumulate within the normal lacustrine sediments. By measuring temperature and chemical variation through the water column and the light penetrating the lake ice a model for a reverse convection cell operating within the lake was devised, and this explains how such a deep lake can maintain a very homogeneous water temperature and chemistry. Finally with the accumulated information a model for present and past proglacial lacustrine sedimentation was developed and this was used as the basis for a proposed facies model for paleo-proglacial lacustrine environments.
|
|---|