Discrete drainage systems are “efficient transporters of meltwater, allowing rapid flow through well connected channel systems” (Benn and Evans 1998). In contrast, distributed systems are “inefficient, and meltwater follows more tortuous routes through poorly connected networks” (Benn and Evans 1998). Therefore the type of drainage has important implications for glacier motion, by controlling water pressure condition at the bed.
A study of discrete subglacial channels was conducted by Seaburg et al. (1988) using dye tracing at Storglaciaren, Sweden. Dye was poured into a ‘Moulin’ in the lower part of the ablation area in the glacier, and the time for the dye to leave the system by way of a meltwater stream was recorded. They found that there was a significant amount of time between when the main cloud of dye came out to the time no dye was seen in the water. The hypothesised reason for this lag time was due to “dispersion by way of a system of discrete subglacial passages” (Bahr, D. B, 1997). Therefore water in subglacial channels flows in response to atmospheric pressure where it is open to the atmosphere, as well as to cryostatic pressure where it is closed off from the atmosphere.
It should also be questioned as of the various routes the water may take once in the glacier, as there are a variety of different ways water can exit the subglacial system. The first was designed by a French geologist, Rothlisberger in 1972. He suggested a system of intra-glacial and sub-glacial channels that melted into the ice by using the processes of viscous dissipation. These were known as ‘R-channels’ and formed “arborescent networks” (Lliboutry, 1979) from one or two input sources; usually crevasses. This definition of ‘R channels’ must therefore mean ‘Eskers’ are a special type of R-channel. Eskers are defined as “subglacial channels that are seen as ridges on the landscape after a glacier has retreated” (Seaburg et al, 1988). However, Eskers do flow from areas of high potential to low potential by the means of viscous dissipation, therefore fitting the criteria as an ‘R Channel’.
The second major group of channels where put forward by Nye (1973) known as ‘Nye channels’, although his N-channels were formed in bedrock irregularities rather than being melted into the ice. The existence of such channels is clearly demonstrated by incised river channels exposed by the retreat of modern glaciers, and abandoned channels cut in bedrock in areas occupied by former ice sheets. N-channels can occur as single isolated features or in braided networks covering large areas of the bed (Walder and Hallet, 1979).The presence of N-channels implies that erosion, and hence water flow, is consistently focused along the same route. Such consistency in the route of water flow will presumably occur where bedrock topography exerts a strong control on the hydraulic gradient, such as steep sided valleys or rough glacier beds. However, this does not appear to be the case, as Sissons et al (1960) suggests that subglacial meltwater channels cut in bedrock do frequently occur along the axes of valleys, or wind between pronounced high points such as ridges.
The final major channel is that of ‘tunnel valleys’. These are “branching channel systems, which can also be incised into subglacial sediments, and can be very large” (Boulton & Hindmarsh, 1987). Hubbard (1995) argued that such valleys develop to allow the efficient drainage of subglacial aquifers, which would otherwise become unstable as a result of high porewater pressures. The spacing of the valleys is dependant on the amount of surface water or basal meltwater to be discharged, and the permeability of the sediment. The pressure conditions and stability of such channels have been considered further by Alley (1992).
More recent studies recognised the presence of discrete sub drainage cavities at the base of glaciers capable of accommodating water, and appearing to have no need for a thick subglacial water film as suggested by Weertman. Lliboutry (1978) thought otherwise in assuming drainage could perhaps be from gradual, seeping through sediment, and through water filled cavities due to melting from geothermal heat and energy dissipation from the internal deformation process. He proposed that cavities became interconnected in the melt season when pressure was high; then cavities would drain through the sediment.
Discrete subglacial drainage could therefore flow in three major ways; either incised up into the ice (Rothlisberger, 1972), incised down into the rock or sediment (Nye 1973) or incised though rock in large valleys (Boulton & Hindmarsh, 1987).
Distributed Systems:
- Water Film – Between ice and bedrock
- Linked Cavity Network – Between ice and bedrock
- Braided Canal Network – Between ice and sediment
- Porewater Flow – (Darcian Flow) Within sub glacial sediments.
There is a mounting body of evidence that relates linked sub glacial drainage to glacial sliding and surge. Although the distinction between surging and non-surging glaciers is clear in principle, there are unexplained reasons why some glaciers surge and others do not. There are obviously differences between simplified model parameters and the complexity of actual glacier beds, but it is difficult to measure these parameters at the base of actual glaciers.
Bibliography
Books
- Seaburg et al., 1988
- Walder and Hallet, 1979
- Benn and Evans 1998
- Boulton & Hindmarsh, 1987
Web Resources
Journals
-
Iken, A (1981) The effect of subglacial water pressure on the sliding velocity of a glacier in an idealised numerical model. Journal of Glaciation. Edition 27 (97):pp. 407-421.
-
Clarke, GKC (1991) Length, width and slope influences on glacier surging. Journal of Glaciation. Edition 37 (126): pp. 236-246.
-
Lliboutry L, (1979) Local friction laws for glaciers: a critical review and new openings. Glacial Geomormopology Journal. Edition 23 (89): pp. 67-95.
-
Rothlisberger H, Flotron A and Haerberli W (1983) The uplift of Unteraargletscher at the beginning of the melt season - a consequence of water storage at the bed? Journal of Physical Geography Edition 29: pp. 28-47
-
Bahr, D. B, (1997) Global distributions of glacier properties: a stochastic scaling paradigm. Water Resources Research, Edition 33(7), 1669--1679