Introduction
The purpose of this note is to draw attention to the disruption of bedrock in a periglacial environment by a process which does not appear to have been reported previously at this scale. During the course of field work carried out in the austral summer of 1975-76 in the South Shetland Islands (Figs 1 and 2), it was noted that there were associations of hummocks and hollows developed on the surface of the lapillite-tuff bedrock of Byers Peninsula, Livingston Island (lat. 62° 40' S., long. 61° 00' W.). The hummocks are in some respects similar to pingos incorporating bedrock (Reference MüllerMüller, 1959) but they are smaller and consist entirely of bedrock. Pingo collapse produces a mound whereas the end product of the cycle of growth and decay of the hummocks described here is a hollow.
Form Morphology and Lithology
The hummocks and hollows were of similar sizes, being approximately 3 m in length 2.5 m in depth and with a relief amplitude of 0.3 m (Table I). The surface of each hummock had a crack running parallel to the long axis of the form and sometimes another at right-angles to it (Fig. 3). The ground surface was covered by a layer of randomly arranged local bedrock clasts and was underlain by a layer of fines 2-6 cm in depth. Beneath this was fractured bedrock (Fig. 4). Excavation of a hummock revealed an ice lens at 0.45-0.6 m below the surface, with a maximum thickness of 0.6 m. The ice was largely debris-free except for its margins where some bedrock fragments were in part incorporated. Vertically aligned lines of bubbles suggest that it is a lens of segregation origin (Reference WashburnWashburn, 1973). Beneath the lens the bedrock was cemented by interstitial ice. The bedrock is a lapillite-tuff with clasts of pumice which are typically between 1 and 10 mm in diameter, though they may be up to 3 cm in size (personal communication from R. E. S. Davies, 1977). The tuff is very friable and breaks along horizontal or sub-horizontal planes to produce platy clasts about 10-15 cm in length, with a maximum of approximately 45 cm.
Environment
The climate of the South Shetland Islands is typical of the maritime Antarctic. Meteorological data are not available for Livingston Island but Reference PepperPepper (1954) has given data for Admiralty Bay, King George Island (lat. 62° 10' S., long. 58° 30' W.) and Whalers Bay, Deception Island (lat, 63° 00' S., long. 60° 30' W.) (Table II). For most years the mean daily temperature is above freezing from December to March, so that melting of frozen ground is possible during this period. Permafrost was found at all sites investigated during the course of field work on Livingston Island. In January and February, when the work was carried out, the depth of the active layer at low altitudes (less than 30 m) was between 0.3 and 0.7 m, depending on materials.
Interpretation
It would seem that the growth of the ice lens is causing the hummock and it is suggested that the hollows result from the collapse of the hummocks following the melting of the ice. A cycle of growth and decay of ice lenses is suggested as follows:
-
1. An ice lens is initiated at a point in the bedrock and begins to grow, heaving up the bedrock which cracks vertically to accommodate the expansion and deformation.
-
2. As the ice lens grows, the sides of the hummock steepen, enhancing mass wasting and resulting in the removal of material from the hummock.
-
3. During the winters the lens increases in size until a threshold is reached when the widening of the crack alters the thermal conductivity of the system to such an extent that the lens is susceptible to melting in subsequent summers.
-
4. With the melting of the lens the bedrock collapses and, because of the removal of material from the system, a hollow is produced.
No climatic change is necessary for the cyclical growth and collapse of ice lenses in this model so that the development of hummocks and hollows is possible under the present climate of Byers Peninsula. Lapillite is the dominant bedrock type of eastern Byers Peninsula, comprising 70% of the succession, so that this process is at least locally important and may be active in other periglacial environments with a suitable lithology.
Acknowledgements
The research was part of a geomorphological programme in the Antarctic supported by the Natural Environment Research Council and the British Antarctic Survey, with study facilities at the Department of Geography, University of Aberdeen. Thanks are due to R. V. Birnie, J. D. Hansom, and Dr J. E. Gordon for assistance in the field, to Dr A. M. D. Gemmell and Dr D. E. Sugden for their comments during the preparation of the manuscript, and to R. E. S. Davies for identifying geological specimens.