Geomorphological expression of glacial erosion or lack thereof

The years following World War II saw earth scientists exploring a wide range of glacier environments from Antarctica to mountain glaciers and ice caps lying at the Equator. This geographical coverage combined with recruitment into the new field of 'glaciol-ogy' of physicists and mathematicians led to the development of models of temperature conditions at the bed of glaciers and the large ice sheets. Sugden (1977, 1978) wrote two seminal papers that first developed an estimate for the temperature at the bed of a North American Ice Sheet (the ice sheet was not temporally fixed), proposed how this might affect glacial erosion or glacial protection, and then used maps and aerial photographs to produce a map of the bed of the LIS showing areas of scour, selective linear erosion (e.g. fiords), alpine glaciation, and areas showing little evidence for glacial erosion. As a dramatic example of the protective nature of an ice sheet where the bed is frozen there is a thin (few centimetres thick) Paleocene (60-70Ma) deposit sitting on a hilltop a few tens of kilometres from the Barnes Ice Cap (Andrews et al., 1972), which in turn is a relict of the LIS (Hooke & Clausen, 1982)!

In terms of the work of Bell & Laine (1985) a key feature of Sugden's map, and subsequent studies (Andrews et al., 1985a,b), was that the 100 m of erosion for the eastern sector of the LIS must have varied spatially from ca. 0 to >>100 m. Because large areas of the eastern Canadian Arctic show little evidence of glaciation (Sugden, 1978) then the rate of erosion in the areas of scour or selective linear erosion must be moderately high. However, it would be a mistake to adopt another of White's arguments (White, 1988) and conclude that the deep sounds and channels that are a feature of the eastern and northeastern margin of the LIS (e.g. Hudson Strait, Frobisher Bay, Cumberland Sound, Lancaster and Jones sounds) are the product of long-term glacial erosion. Undoubtedly ice streams have flowed along these features but in all these cases there is undisputable geological evidence that they are fault-bounded and, moreover, have 'soft' sedimentary rocks lying at the seafloor (MacLean, 1985; MacLean et al., 1986; MacLean, 2001a,b), these would include mainly Palaeozoic carbonates but in the case of Cumberland Sound, Cretaceous mud-stones. Thus the large-scale features that fringe the margin of the LIS and often lead into its interior are grabens, probably date to the Tertiary break-up of Canada and Greenland. This being the case then the question has to be revisited—what are the origins of the fiords that are such a dramatic feature of the Labrador and Baffin landscape, or, more specifically, what fraction of their current volume is associated with the removal of rock by glacial erosion and how much might be allocated to the creation of space by faulting (Dowdeswell & Andrews, 1985)? This is an unanswered question and probably both processes explain some fraction of the fiord volumes. At ODP Site 645, off the coast of Baffin Island and on the floor of Baffin Bay (Hiscott et al., 1989; Srivastava et al., 1989), the rate of sediment accumulation is ca. 0.13 mkyr-1 and the rate of accumulation at other sites within Baffin Bay are also relatively modest at ca. 0.1 mkyr-1 (Andrews et al., 1998a), especially given the fact that the Bay was, and is, surrounded by large ice sheets and glaciers.

How much erosion has occurred on the scoured landscapes of the Canadian Shield is difficult to estimate but along the eastern sector it is worth noting that the erosional products from the scoured areas have to transit the uplands, which show little evidence of active glaciation (Andrews et al., 1985a), before converging into fiord outlet glaciers and in some cases ice streams. On the Baffin margins these ice streams are marked by plumes of carbonate-rich till (Tippett, 1985), which indicate transport from the Palaeozoic outcrop around and in Foxe Basin toward the eastern coast of Baffin Island (Fig. 40.3). Suprisingly, the major trough that leads to Cumberland Sound does not contain such a plume (Andrews & Miller, 1979).

In other parts of the LIS field observations and glaciological modelling indicate that the LIS, like the Antarctic and Greenland ice sheets today, contained fast-flowing ice streams (Marshall & Clarke, 1996) that characterized the southern, eastern and northern margins. As in Baffin Island the paths of these former ice streams can be delineated by bedforms and frequently by the composition of tills, often detectable from satellite imagery (Dyke & Morris, 1988), especially when carbonate-rich tills are spread across the rocks of the Canadian Shield. Other distinctive rocks are used for tracers of glacial transport, however, such as the red bed sandstones of Keewatin that were transported toward and along Hudson Strait (Shilts, 1980; Laymon, 1992; Aylsworth & Shilts, 1991) and a unique volcanic rock from the Belcher Islands, Hudson Bay, that Prest (1990) traced for thousands of kilometres to the south and west toward the LIS margin (Fig. 40.1).

Fisher et al. (1985) modelled the impact of a deforming bed on the shape and volume of the LIS. The notion of a deforming bed was then quite new and was based in part on measurements from beneath Icelandic glaciers (Boulton & Jones, 1979) and theory and observations from Antarctic ice streams (Hughes, 1977; Alley et al., 1994; Weertman & Birchfield, 1982). There are two geographical areas of deforming beds under the LIS—one is linked with the ice streams that follow the structural channels around the eastern and northern margins of the ice sheet (e.g. Hudson Strait, Cumberland Sound, Lancaster Sound, etc), whereas the broader concept involves the ring of softer sedimentary rocks that lie to the south and west of the Precambrian shield (Fig. 40.3) and White's (1972) arc of exhumation (but see Stokes & Clark, 2003a). Clark & Walder (1994) showed that the distribution of eskers at the bed of the LIS is largely confined to the hard-bed of the shield, whereas on the fringing area of sedimentary rocks (Fig. 40.3) there are few eskers, but it is in this area where Mathews first identified outlets of the LIS with extremely low gradients (Matthews, 1974), and this work has been extended along this fringe. The presently accepted notion is that these outlets were lying on deforming sediments (Clark, 1994; Clark et al., 1996b) but there are challenges to this inclusive notion (Stokes & Clark, 2003a).

One final aspect of issues pertaining to the rates of glacial erosion is the tremendous potential for the application of cos-mogenic exposure age dating to not only issues of glacial chronology but also to the fundamental question of the rate of glacial erosion. Near the southern margin of the LIS Colgan et al. (2002) collected 22 samples from five striated rock outcrops comprising granites, metarhyolites and quartzite. In two of the outcrops nuclide abundances were consistent with accumulation since deglaciation, implying an erosion of ca. 2 m or more of rock. However, in three outcrops the nuclide abundances '. . . were up to eight times higher than predicted by the radiocarbon chronology'. For these data minimum limiting glacial erosion rates of 0.01-0.25mmyr-1 were estimated (Cogley et al., 2002, p. 1581). Clearly it will be extremely interesting to obtain cosmogenic exposure age dates from the unscoured uplands of Baffin Island and Labrador—this is currently being investigated (Briner & Miller, personal commununication, 2004).

Was this article helpful?

0 0

Post a comment