Other Marine Glaciers

The ice front of three other ice shelves in the Pacific Sector have retreated in recent years. The fronts of the Wordie and George VI Ice Shelves on the west coast of the Antarctic Peninsula have retreated by several kilometres over the last 10 to 30 years (Doake, 1982). About one-third of the Wordie Ice Shelf has disappeared. This behaviour may be due to a warmer climate and decreasing snowfall in recent years, although glaciers in parts of the Peninsula are still thickening apparently in response to a high accumulation rate experienced several hundred years ago. It is, therefore, not possible to attribute the retreat unambiguously to disintegration of the ice sheet in the Antarctic Peninsula (Doake, 1985). Similarly, the retreat of the northern margin of the McMurdo Ice Shelf in the south-west Ross Sea, shown in maps by, for example, Prebble (1968) and Pyne et al. (1985), may be due simply to short-term disequilibrium between ice input, bottom melting, and the rates of calving. The Nansen Ice Shelf in western Terra Nova Bay is presently advancing after a calving event between 1957 and 1973 removed about 15% of the ice shelf area, as indicated by comparing photographically derived maps and Landsat imagery. These four ice shelves exist near the limits for their climatic and oceanographic environments and should be monitored for possible response to climate change (Mercer, 1978; Doake, 1982).

The dynamic behaviour of these ice shelves is different from that of Ross Ice Shelf. Descriptions in Swithinbank (1970) and Doake (1982, 1985) suggest that brine percolation, surface and bottom melting, and calving are more significant relative to the size of the ice shelf than in the case of the large ice shelf. Ice velocities are also less than on the Ross Ice Shelf (Heine, 1967; Swithinbank, 1970; Doake, 1985; J.R. Keys, unpubl. data).

The Erebus Glacier Tongue is the most studied ice tongue in the Pacific Sector. Holdsworth (1978, 1982) has described its morphology, dynamics and possible modes of calving. The tongue, currently 14 km long (1988), thins from 300 m thick and 2.8 km wide at the grounding line on Ross Island to 60 m thick and 1 km wide at its snout. Surface densities vary from that of snow to ice (at 0.87 ± 0.02 g cm-3) while surface mass balance is calculated to be ± 0.1 m yr1 along the tongue. Bottom melting appears to predominate and derived rates are up to 4 myr1 on the base and 10 m yr1 on the sides. Ice velocity near the centre-line increases from

90 m yr1 near the hinge-line where the ice flows off Ross Island, to 150 m yr1 at the snout. The tongue underwent its most recent major calving in the early 1940s and has been advancing at an accelerating rate since then, with speeds similar to the ice velocity at the snout (from Holdsworth, 1985).

Various maps and recent observations of the snow-covered Drygalski Ice Tongue in the western Ross Sea suggest that this 70 km long, 20 km wide tongue has advanced in recent years at a rate similar to the speed of its ice (730 m yr1) about 50 km from the snout) (Holdsworth, 1985). Up-warping of floating "capes" is prominent along the northern margin of the tongue, indicative of strong wave action or surface currents causing ablation near sea level to exceed melting at depth.

The tongue of the Thwaites Glacier in the Amundsen Sea has undergone complex behaviour in recent times. Visual imagery from Kosmos-226 satellite suggests that the glacier tongue apparently doubled in length between 1961 and 1969 to about 200 km, but also broke off and buckled laterally towards the west in this period (Savatyugin, 1970; Holdsworth, 1985). A glacier tongue 70 km long was present in 1972 with an iceberg tongue 130 km long immediately to the north. This iceberg tongue appears to be rotating anticlockwise about a probable pinning point roughly 200 km from the glacier's grounding line (Hughes, 1983). Glacier flow and the west-setting Antarctic coastal current (East Wind Drift) may be causing this rotation. The glacier tongue has advanced since 1972 apparently at a rate similar to the surge-like speed of its ice near the ice front (3.7 km yr1) (Lindstrom and Tyler, 1984).

Complex, dynamic behaviour of the Thwaites Glacier and its neighbour, the Pine Island Glacier, in the Amundsen Sea is not inconsistent with the theories of Hughes (1973,1983) linking these glaciers to a drop in the level of their drainage basins and collapse of the West Antarctic Ice Sheet. These fast-moving glaciers also seem to have high basal melt rates which might accelerate decay of ice shelves formed in the embayment there or prevent them forming at all. However, radio echo sounding shows that the grounding line of the Pine Island Glacier is held at the foot of a rock bar 200 m high which, together with simple modelling, suggests that the glacier is stable and not undergoing any unusual behaviour (Crabtree and Doake, 1982; Doake, 1985). This also applies to the Thwaites Glacier (Mclntyre, 1985).

Nevertheless, neither glacier is necessarily in equilibrium. If either of their grounding lines began a rapid retreat, they could accelerate their drainage and possibly destabilize the West Antarctic Ice Sheet. Modelling has suggested that it could take a few hundred to a few thousand years for the Ice Sheet to collapse in this way (Anonymous, 1985a), but estimates are tentative because of the lack of knowledge of the dynamics of fast-moving ice streams (Robin, 1986).

0 0

Post a comment