AbstractThermally induced seasonal movements of the active layer and subjacent permafrost have been measured in numerous ice-wedge polygons that have varied in age, type, crack frequency, and topographic location. The field observations show that, in winter, thermal contraction, which is inward, is constrained or vanishes at the polygon centres but, in summer, thermal expansion, which is outward, is unconstrained at the ice-wedge troughs. Therefore, there tends to be a small net summer transport of the active layer, to varying depths, into the ice-wedge troughs. The movement has been observed in all polygons studied. The slow net transport of material into the ice-wedge troughs has implications for: permafrost aggradation and the growth of syngenetic wedges in some troughs; the palaeoclimatic reconstruction of some ice- wedge casts; and the interpretation of polygon stratigraphy based upon the assumption that the polygon material has accumulatedin situ.

No abstract

Underground ice is restricted to permafrost areas where its distribution is sporadic and often unpredictable.

Growth data from precise surveys have been obtained for 11 pingos for periods ranging from 20 to 26 years. Most of the 1350 pingos, perhaps one quarter of the world's total, have grown up in the bottoms of drained lakes underlain by sands. Permafrost aggradation on the drained lake bottoms has resulted in pore water expulsion, solute rejection below the freezing front, a freezing point depression, and groundwater flow at below 0° C to one or more residual ponds, the sites of pingo growth. Sub-pingo water lenses underlie many growing pingos.The pure ice which grows by downward freezing in a sub-pingo water lens may be composed of seasonal growth bands which, like tree rings, are of potential use in the study of past climates. Growing pingos underlain by sub-pingo water lenses can often be identified by features such as peripheral pingo rupture, spring flow, frost mound growth, normal faulting, and oscillations in pingo height. Such features, and others, are associated with hydrofracturing and water loss from a sub-pingo water lens. Some of the data derived from the long-term study of pingo growth are relevant to the identification of collapse features, interpreted as paleo pingos, in areas now without permafrost.

Field studies carried out mainly since 1975 in permafrost areas of Alaska, Canada, China, and the Soviet Union have been combined with the results of laboratory investigations to show that in summer water can move from the thawing active layer into the subjacent frozen active layer and under certain conditions even into the top of permafrost. Direct field evidence discussed includes: data from drilling and neutron probe logging, which show a summer increase in the ice content of already frozen ground; summer heave of heavemeters, with heave occurring in the frozen active layer; and increase in the ice content of the subjacent frozen ground in both permafrost and non-permafrost areas, caused by snowmelt infiltration. Indirect field and laboratory evidence is also added to support the direct lines of evidence. The conditions that favor the downward migration of water from thawed to frozen ground are examined in terms of thermally induced hydraulic gradients, hydraulic conductivity, content of unfrozen pore water, temperature gradients, ice content, and gravity. Some geocryologic implications of the summer growth of ice in frozen ground, including the effects on water balance calculations and the origin of patterned ground, are briefly mentioned.