B

Asymmetric and symmetric mounds (Type I)

Small-scale mounds (Type I transitional to II & IV)

^^ Ridged mounds (Type III and IV) \\ Longitudinal mounds (Type V)

# Moraine Plateaux (Type VI) Linear ridges (moraines?) Eskers

Meltwater channel boundary

Figure 8.3 (a) Aerial photograph from the McGregor Lake Reservoir region. (b) Interpretation of ridges on aerial photographs based on photograph analysis and field work. Small map inset at base shows the position of the photograph relative to the area presented in Fig. 8.2.

Asymmetric and symmetric mounds (Type I)

Small-scale mounds (Type I transitional to II & IV)

^^ Ridged mounds (Type III and IV) \\ Longitudinal mounds (Type V)

# Moraine Plateaux (Type VI) Linear ridges (moraines?) Eskers

Meltwater channel boundary

Figure 8.3 (a) Aerial photograph from the McGregor Lake Reservoir region. (b) Interpretation of ridges on aerial photographs based on photograph analysis and field work. Small map inset at base shows the position of the photograph relative to the area presented in Fig. 8.2.

48 J. Shaw & M. Munro-Stasiuk preted by Allen (1982) as products of turbulent, separated flow. It is literally impossible for laminar flow to have transported the boulders and formed the erosional marks. Contrary to the opinion of Benn and Evans, Reynolds number calculations follow from the field observations; the flow depths and velocities required for these calculations are not assumed.

Benn and Evans provide an extensive discussion on the source and amount of water for the floods. They miss the point. We do not argue that melt-out produces all the water for floods. Munro-Stasiuk (2000) specifically noted that the melt-out processes (associated with small volumes of water) preceded the flood events and that the water for the major flood was part of the Livingstone Lake Event and was derived far to the north of her study site. She also provided extensive evidence for sedimentation into subglacial reservoirs that occupied the local pre-glacial valley system (Munro-Stasiuk, 2003). The sedimentary facies representative of the reservoirs consists of a range of subaqueous deposits (mostly gravity flow deposits and not melt-out deposits) that are chaotically deposited and range from entirely undisturbed to pervasively deformed (interpreted as ice-reactivation). Although Benn and Evans go on at length arguing that the subglacial reservoirs discussed by Munro-Stasiuk (2003) were not large enough to provide the volume of water required for a megaflood, Munro-Stasiuk noted that the reservoirs were 'small', never contended that the reservoirs were the source of the megaflood waters, and clearly stated that the presence of reservoirs was followed by till deposition and then by the event responsible for creating the ero-sional landforms. In this volume, Shaw (Chapter 4) refers to abundant meltwater to refute the notion of frozen bed conditions near the centre of the ice sheet and to account for hydraulic connectivity between the central zone of the ice sheet and the margins. Shaw (1996) proposed a supraglacial origin for the melt-water, a suggestion that is well supported by observations on modern glaciers and by modelling of past ice sheets (Marshall et al., 2002; Zwally et al., 2002a). This proposal clearly establishes that melt-out is not considered as the primary source of meltwa-ter for the flood events.

Benn and Evans argue that modern glacial environments provide all the necessary analogues for past glacial landscapes. This claim is unlikely. For example, there are no known modern glacial landforms resembling Rogen moraine, nor any that show the sedimentary and morphological characteristics of hummocky moraine on the Western Plains of Canada. Furthermore, the streamlined, loess hills of the unglaciated Channelled Scablands in Washington are identical in form to drumlins 100 km or so to the north. Also, the fluted bedrock above Dry Falls in the Scab-lands is identical to similar bedrock fluting in glaciated areas. Both of these Scabland landforms required immense water sheet floods for their formation. They were formed beyond the ice limits and it is therefore impossible that these could have been created by glacial processes. At the same time they provide powerful analogues for drumlins and erosional marks. Obviously, we need something other than analogues from modern glacial environments to explain these non-glacial landforms. Thus, it is reasonable to use Scabland flood landforms as part of a hypothesis on subglacial megafloods.

Shaw never claimed that drumlins and fluting do not show cross-cutting relationships. In fact, Shaw (1996, fig. 7.39) presents a map clearly showing such cross-cutting. We cannot imagine why megafloods could not cause cross-cutting. All that is required is a variation in flow direction. Indeed, Shaw & Gilbert (1990) differentiated the Algonquin and Ontarian events on the basis of the cross-cutting of one flow path by the other. The Algonquin came first. By contrast, there are vast (hundreds of kilometres long) tracts of drumlins and flutings, the Livingstone Lake flow path for example, that do not show cross-cutting relationships, nor do they display systematic superimposed forms. It is our view that these tracts represent pristine subglacial surfaces.

In summary, Benn and Evans state that the megaflood case 'is shown to rely exclusively on the perceived morphological similarity between drumlins and streamlined forms eroded by turbulent flows'. As discussed above, the megaflood case is grounded in sedimentary, hydrological and glacial theory, and is supported by extensive morphological and sedimentary observations and interpretations.

Was this article helpful?

0 0

Post a comment