We have reviewed the broader aspects of ice-marginal landform distribution in the continuous permafrost zone of northern Canada. The environment places constraints on possible basal ice thermal conditions in glacier marginal and sub-marginal zones throughout glacial cycles. Ice caps necessarily nucleated and expanded on pre-existing permafrost. The permafrost then either survived throughout the glacial cycle or reformed upon or prior to deglaciation. Ice-marginal landforms and bedforms in this region are zonally distributed along the ice sheet radius in a manner that can be interpreted in terms of the relative prevalence of cold-based and warm-based ice. Where the ice sheet was extensively cold-based, ice-marginal landforms are nearly limited to lateral and proglacial meltwater channels and ice-thrust moraines. The latter are most common and most voluminous where composed of disturbed unconsolidated sediment. Broad constructional morainal belts are extensively cored with glacier ice and backed by streamlined terrain. These belts formed along a cold-based marginal fringe backed by warm-based, debris-rich ice. Characterized by large kettles, these morainal belts can be misinterpreted as thermokarst terrains. They would lose much of their topographic organization and misleadingly resemble dead-ice terrain were they to lose their ice cores. Short of serious global warming, loss of ice cores is improbable in the continuous permafrost zone. Such ice-cored terrain probably survived through previous interglacials and likely provided large, regional debris sources to advancing cold-based ice during phases of glacial build-up. Where the ice sheet was extensively warm-based, few regionally significant end moraines formed except during readvances. Rogen moraines probably formed from previously deposited drift along the boundary between distal, sliding and proximal, non-sliding (cold-based) ice and hence are well preserved only near final ice recession centres.
We know of no extensive terrain south of permafrost in North America that resembles the northernmost terrain outlined above with its extensive marginal meltwater channels. Less-extensive occurrences have been described from the Cordillera (Tipper, 1971; Dyke, 1990) and from central Quebec-Labrador (Ives, 1960). However these are areas of discontinuous permafrost today, and it is therefore probable that the marginal zone was cold-based during deglaciation.
Considerations of former permafrost conditions during deglaciation have already been thoughtfully incorporated into conceptual models of landform genesis along the southern margin of the Laurentide Ice Sheet starting perhaps with the seminal model of Clayton and Moran (1974), which incorporated a cold-based marginal fringe (see Chapter 1). This model was applied to interpreting the landscape zonation of the James and Des Moines lobes of the Laurentide Ice Sheet. The broad morainal belts there can thus be seen as former ice-cored moraines, and the still-ice-cored moraines of similar size in the Canadian Arctic can serve as analogues of the evolutionary stage of the southern moraines after deglaciation but before loss of ice cores. More recently, Attig et al. (1989) and Clayton et al. (2001) have attributed the formation of hummocky end moraine in Wisconsin to compressive flow across a frozen toe zone creating broad belts of ice-cored moraine. The inferred ice-walled lakes within the ice-cored terrain (e.g. their Fig. 1.1) are equivalent to the extant kettles in the moraines of Arctic Canada. Ham and Attig (1996) present a similar model for end moraines further west in Wisconsin. These models are entirely compatible with our observations from the present permafrost zone. However, we would prefer to refer to the forming or detached zone of buried ice, a transient permafrost feature in southern areas but a permanent one in the north, as ice-cored moraine rather than stagnant glacier ice. This is because they formed along active, commonly readvancing, ice margins rather than by regional stagnation.
In the area described by Attig et al. (1989), broad end-moraine belts ceased to form once regional permafrost disappeared and drumlin formation extended to the ice margin. This is a reasonable inference because it is based in part on the distribution of palaeo-permafrost features (Clayton et al., 2001). However, landscape zonation within the permafrost zone described above also shows that retreating ice sheets can be warm-based right out to the margin where permafrost is necessarily forming upon deglaciation.
These examples and more direct evidence demonstrate that permafrost was widespread along the southern ice margins at the last glacial maximum and early in deglaciation. Clayton and co-workers have incorporated permafrost into glacial process-form models, with the cold-based marginal fringe being the key element. We suspect, however, that this model is still under-applied in interpreting landscape zonation in presently non-permafrost regions.
Discussions with Larry Dyke, Geological Survey of Canada, about permafrost evolution through glacial cycles is particularly appreciated. The manuscript was reviewed by Greg Brookes, Geological Survey of Canada, who clarified several aspects.
Was this article helpful?
What you need to know about… Project Management Made Easy! Project management consists of more than just a large building project and can encompass small projects as well. No matter what the size of your project, you need to have some sort of project management. How you manage your project has everything to do with its outcome.