and low annual precipitation receipts of ca. 300 mm. In contrast, the southeast Alaska climate is a cool, wet maritime climate (Figure 7.4). These differences translate into large differences in glaciological regimes, the potential for glacial erosion, and in an order-of-magnitude difference in the volume of meltwater generated by both snow and ice melt (Syvitski et al., 1987a; Powell and Molnia, 1989). An additional major difference between these regimes (Figure 7.4) is that the production of run-off from snow/ice melt and rainfall is restricted to around 90 days in the eastern Canadian Arctic but is potentially year-round in southeast Alaska.
Strong winds are a feature of areas adjacent to ice sheets and glaciers. This, combined with limited vegetation cover, suggests that, in our area of interest, eolian deflation of outwash sediments is a component in highlatitude sediment flux calculations (cf., Gilbert, 1982, 1983; Syvitski and Hein, 1991).
The extent of sea ice (Figure 7.1) and icebergs is extremely different from area to area. The former is related to the air temperature, wind stress, and temperature of the surface layer of the shelf watermass. The size and durability of icebergs is also a matter of air temperature, wind regime, and the character of the source glacier (Dowdeswell and Murray, 1989). Icebergs from fast-moving tidewater glaciers are relatively small, especially compared with the massive icebergs that are produced from outlet glaciers of the Greenland Ice Sheet. A landfast sheet of sea ice during the winter months (Figure 7.1) serves to buffer glacier margins from wave attack, reduces the calving rate, and restricts iceberg passage for several months per year.
Glaciology and Hydrology The critical aspects of glaciology and hydrology for marine sediment flux studies are the mass balance gradient, the thermal conditions at the ice surface and base, the percentage of the basin that is glacierized, and the location and shape of the terminus — does it end in tidewater or on land, and does it have ramp or shelf, or end as a near vertical face? The presence of ice in a drainage basin can extend both the volume of meltwater discharge, by amounts proportional to the area covered by ice, and may also extend the duration of nival and ice associated runoff. However, the origin of sediment carried in the meltwater streams depends on the thermal conditions of the ice body. In temperate areas, with MATs close to 0°C, surface run-off is produced in summer by ablation; this water drains into the glacier and moves toward the ice front in englacial or subglacial conduits. At the base of the glacier, it adds to water already at the bed which has been produced by melting associated with the geothermal heat flux and the frictional heat generated by the ice sliding over its bed. These combined sources result in an annual melt rate at the bed of between <1 mm to several centimeters (Drewry, 1986). By contrast, subpolar environments with MATs << 0°C have meltwater generated on the surface which runs off in supraglacial stream channels on the ice surface. If the supraglacial stream reaches a lateral margin of a glacier it will run along the glacier in a marginal or submarginal drainage channel. Thus, in a temperate glacier, the meltwater reaches the bed of the glacier and can remove fine-grained sediment produced by glacial abrasion and crushing. However, in subpolar glaciers the surface streams are either very clean (if they run over the glacier) or turbid if they can erode sediment along the side of the glacier.
Because of ice convergence into fiords, it is rare to find outlet glaciers reaching the sea that are not at the pressure melting point (PMP) at their beds. Thus, most tidewater glaciers probably discharge water into the sea from basal melt, which rises as a freshwater plume (Pfirman, 1985). Depending on the sediment load, some sediment may be transported via underflows or interflows.
In fiord basins dominated by meltwater there are significant differences between regimes where the fiord is occupied by one or more tidewater glaciers, compared to the situation where the glaciers terminate on land (Syvitski et al., 1988; Syvitski, 1989). In the first situation, sediment deposition is controlled by (1) processes at the ice front (sub-and englacial-meltwater discharge; the flow of supraglacial till, etc.); (2) hemipelagic sedimentation from the fluvial plume; (3) sediment rafting by icebergs and sea ice; and (4) deep-water currents. In contrast, when glaciers end on land, the dominant sediment supplies are from: (1) bedload dumping from the delta front; (2) hemipelagic sedimentation under the seaward flowing river plume; (3) proximal slope bypassing by turbidity currents; and (4) the combined effects of both short- and long-term downslope diffusion of the accreting sediment mass. There is a significant difference in the grain-size distributions between these two cases for the suspended sediment plume (Pfirman, 1985). This is related to the energy loss from a rapidly rising plume (from a subglacial position) compared to the normal fluviodeltaic discharge to the fiord head (Syvitski, 1989). Thus, for a given level of discharge, coarser sediments are deposited closer to the tidewater glacier, whereas equivalent size particles can ride the freshwater surface plume seaward in the fluviodeltaic fiord.
A third glaciological situation exists in true polar climates (i.e., Antarctica), where even in summer no surface melting takes place; basal melting is possible only if the glacier is sliding over its bed.
In the discussion above, we have implicitly assumed that the bulk of glacial sediment is being delivered to the marine environment via meltwater. An alternative delivery system has been proposed, based on theory and limited observation (cf., Boulton and Jones, 1979; Blankenship et al., 1986). This suggests that a glacier may lie on a deformable sediment bed which is transported seaward by