Box 85 Sediment Transport In Glacial Meltwater Streams

Reliable estimates of bedload transport within meltwater streams are difficult to obtain, due to the high flow magnitudes and large sediment volume involved. One of the first studies to provide reliable estimates of bedload transport was that of 0strem (1975). Two techniques were used to measure the bedload of a proglacial stream in front of the glacier Nigardsbreen in Norway.

The first method involved the construction of a 50 m steel fence across the main meltwater stream during the summer of 1969. The mesh size was such that it trapped all sediment larger than 20 mm in diameter. The accumulation of coarse bedload trapped by the fence was measured by probing the depth twice a day at 176 points along the fence.

The fence survived for 3 weeks before being destroyed in a flood. Between the 24 May and 19 June, 400 tonnes of material were trapped. Samples of meltwater were also taken during this period to determine the suspended sediment content. During this period approximately 1200 tonnes of suspended sediment were discharged. This suggests that during the study period bedload transport accounted for 25% of all the material transported, although it must be noted that sediment in the size range of 1-20 mm in diameter was not trapped and therefore this value must be regarded as a low estimate.

The second method involved the annual survey of a delta formed as the meltwater from Nigardsbreen enters a lake about 1 km in front of the glacier. Most, if not all, of the bedload in transport is deposited on this delta. The average annual accumulation on this delta is 11 200 tonnes, giving a crude estimate of the total bedload moved each year by the meltwater streams.

In recent years the sophistication of bedload traps has improved and accurate estimates of bedload transport are more common place. However, recent work simply confirms the high levels of bedload transport within meltwater streams first quantified in detail by 0strem.

Source: 0strem, G. (1975) Sediment transport in glacial meltwater streams, in Glaciofluvial and Glaciolacustrine Sedimentation (eds A.V. Jopling and B.C. McDonald), The Society of Economic Paleontologists and Mineralogists, Special Publication 23, pp. 101-22.

1. The proximal zone. In this zone sedimentation is dominated by: (i) the changing position and geometry of the ice margin and/or any ice-cored ridges present (Figure 8.15); (ii) the rate of supply or availability of supraglacial meltout till; and (iii) the seasonal flood regime. Braided stream flow is only one part of the total hydraulic system. Resedimentation of supraglacial meltout till as mud, debris and subaqueous flows is common and may dominate the depositional processes. The availability of large quantities of readily transported sediment and the rapid build up and decay to and from flood discharges has a strong effect on the character of the fluvial sediments deposited. Meltwater streams are often incompetent to transport the available till load and may simply redistribute it as structureless, matrix-supported, outwash unit in which the particle-size distribution is little different from the parent till. During flood phases all available particle sizes may be transported and deposited simultaneously. Weak stratification at the top of these massive units of outwash may develop during the waning flood stage and consists of individual lamellae or sediment layers several clasts thick formed by the removal of the fines (winnowing) to give an armoured, often scoured, surface. Proximal outwash is frequently found inter-bedded with units of flow till and other mass-flow deposits, particularly where melt streams are bordered by ice-cored debris ridges. Consequently, the massive unstratified proximal outwash typical of this zone is difficult to distinguish from supraglacial meltout till.

Deposition also frequently occurs on buried ice, the meltout of which causes deformation or subsidence structures. These usually consist of normal or extensional faults and synclinal fold or sag structures (Figure 8.18). If melt-out and subsidence occurs while deposition is still taking place these subsidence structures are referred to as syn-sedimentary. Evidence for subsidence may not be present on the surface, because meltwater deposition is concentrated in the subsiding areas and infills them (Figure 8.19). If meltout occurs after deposition has finished then the landsurface is deformed and parallels the bedding (Figure 8.19). These subsidence structures are post-sedimentary (see Section 9.3).

Subsidence Deposition

Figure 8.18 Faulted sands within outwash sediments in front of Skeidararjokull, Iceland. These normal faults formed during subsidence associated with the meltout of buried ice. [Photograph:

Figure 8.18 Faulted sands within outwash sediments in front of Skeidararjokull, Iceland. These normal faults formed during subsidence associated with the meltout of buried ice. [Photograph:

240 Glacial Sedimentation on Land A Syn-sedimentary subsidence

2. The medial zone. Away from the ice margin a braided river pattern develops, in which ephemeral bars and channels dominate (Figure 8.20). Individual channels and bars vary in size from a few metres to hundreds of metres in width. The depth of channels is typically only a few metres. The braided pattern develops in response to the large sediment load, high discharge and the steep gradient of outwash surfaces. The particle size of sediment deposited rapidly falls away from the glacier margin, which reflects the decline in stream power and the

Figure 8.20 Braided outwash plain, Entujokull, Iceland. [Photograph: M. R. Bennett]

processes of clast attrition in highly turbulent channels (Figure 8.21). Three types of bar or sandwave can be identified, although they are not unique to glacial outwash systems (Figure 8.22).

Figure 8.21 Relationship between clast size and distance from the glacier margin for a variety of glacial outwash systems. [Modified from: Drewry (1986) Glacial Geologic Processes, Arnold, figure 10.10, p. 159]

Figure 8.21 Relationship between clast size and distance from the glacier margin for a variety of glacial outwash systems. [Modified from: Drewry (1986) Glacial Geologic Processes, Arnold, figure 10.10, p. 159]

Linguoid Bar
Figure 8.22 Schematic illustration of the three main types of bars found in braided outwash channels. [Modified from: Drewry (1986) Glacial Geologic Processes, Arnold, figure 10.11, p. 160]

A. The longitudinal bar. This forms in a mid-channel position when the coarsest part of the stream load is deposited as stream flow falls and loses competency. These bars are orientated roughly parallel to the current flow. They are small when first formed, but continue to grow in length and height as fine sediment is deposited downstream in the lee of the bar. Grain size tends to decrease downstream.

B. The linguoid or transverse bar. This type of bar is orientated transverse to the direction of stream flow and may possess a diamond or rhombic shape with a steep downsteam face. They develop under high flow conditions and extend by particles avalanching down their front face.

C. The lateral, side or point bar. These bars are typically very large and develop on the sides of stream channels in quieter water. They are attached to the sides of the channel.

With distance from the glacier margin the proportion of linguoid or transverse bars increases as does the sand component within the fluvial system. Sediments resulting from this braided channel system range from boulders to sands. The sedimentary structure within these sediments reflects five processes that operate in the braided channel system. These are: (i) the formation of bars; (ii) the formation of bedforms such as ripples and dunes in finer material; (iii) the erosion (scour) and fill of channels; (iv) the deposition of finer sediments during low flows, particularly in backwaters; and (v) overbank sedimentation during the falling stage of flood flows. Table 8.3 shows the types of sedimentary facies associated with these processes.

Table 8.3 Diagnostic criteria for recognition of common glaciofluvial stream deposits.





Massive, matrix-supported gravel with no sedimentary structure

Debris flow


Massive or crudely bedded gravel with horizontal

Longitudinal bars and channel

bedding and clast imbrication

lag deposits


Stratified gravel with trough cross-beds

Minor channel fills


Stratified gravel, planar cross-beds

Transverse or linguoid bars


Medium to very coarse pebbly sand with solitary or grouped trough cross-beds



Fine to very coarse pebbly sand with solitary or grouped planar cross-beds

Transverse or linguoid bars


Very fine to coarse sand, with ripple marks

Ripples, low-flow regime


Fine to very coarse often pebbly sand, with horizontal

Planar bed flow, high-flow




Fine to coarse sand, may be pebbly, with broad shallow scour structures

Minor channels or scour hollows


Sand, silt and mud, with ripple marks

Waning flood deposits and overbank deposits


Mud and silt with desiccation cracks

Drape deposits formed in pools of standing water

For a detailed description of the types of cross-bedding referred to in the table the reader is referred to the volume by: Collinson, J.D. and Thompson, D.B. (1989) Sedimentary Structures. Unwin Hyman, London.

[Modified from: Eyles et al. (1983) Sedimentology, 30, table 1, p. 395]

3. The distal zone. The proportion of fine-grained sediment increases dramatically with distance from the glacier margin. During normal discharge the main flow is concentrated in a single channel, although at peak discharge a braided pattern may develop. The glacial influence decreases and the flow regime is dominated less by the seasonal patterns of ice melt. There is a gradual transition into a more conventional fluvial system.

The pattern of sedimentation in the proximal, medial and distal zones will be disrupted if the glacier is subject to catastrophic floods or jokulhlaups. Sedimentation will be partly controlled by the shape of the flood hydrograph, which is a function in part of the cause of the jokulhlaup (see Figure 4.12). Floods triggered by volcanic activity beneath a glacier are typically of high magnitude but of short duration, in contrast to floods caused by the drainage of ice-dammed lakes, which tend to be of longer duration. Research to date suggests that jokulhlaup discharges result in sediment sequences consisting of massive, poorly sorted, non-graded or inversely graded sediments. The latter are characterised by large surface boulders, around which flow may be channelled. If present, these channels tend to contain boulder lags, fields of mega-ripples and streamlined boulder hummocks. The main unit is interpreted as the product of hyperconcentrated fluid-sediment mixtures that are sufficiently dense to transport large boulders on their surface and prevent any size-sorting or grading from developing. These hyperconcentrated flows are associated with the initial flood surge. As the flow stage declines the sediment solidifies and dewaters. This water may then form a series of more fluid flows on the surface of the main deposit. If these fluid flows are sufficiently large they may scour channels and deposit the lag horizons. The massive units of boulder-rich gravel typical of jokulhlaups provides a sharp contrast with the deposits produced by normal glacial discharges.

Was this article helpful?

0 0

Post a comment