The Regional Impacts of Climate Change

	Other reports in this collection

(continued...)

Ramsden and Fleming (1995) suggest that Arctic sea ice would be little altered under doubled CO₂ conditions. More recent analyses by Gordon and O'Farrell (1997), who use a dynamic ice procedure within a transient-coupled-atmosphere-ocean model, predict a 60% loss in summer sea ice in the Northern Hemisphere by the time CO₂ has doubled. In their model, ice loss appeared earlier and proceeded more rapidly than it did in the Southern Hemisphere.

Although current polar models are subject to considerable uncertainty, they are the best available tools to explore possible future scenarios. A regional study of the Canadian sector of the Beaufort Sea proposed an increase from 60 days to 150 days in the open-water (ice-free) season (McGillivray et al., 1993). There could also be reduced ice thickness. The maximum extent of open water in summer could increase from the current 150-200 km to 500-800 km offshore. At the same time, if longer fetches are available, wave heights would increase (e.g., the proportion of waves in excess of 6 m would rise from 16% to 39%) (IPCC 1996, WG II, Section 7.5.4). This increased open-water exposure of the ice-rich coastlines would most certainly result in increased erosion.

Using empirical ice growth-melt models, Wadhams (1990) predicts that in the NorthWest Passage and Northern Sea Route, a century of warming would lead to a decline in winter fast-ice thickness from 1.8-2.5 m at present to 1.4-1.8 m and an increase in the ice-free season of 41-100 days. Other researchers, using another mix of models, find these ice-free day estimates a little high (Flato and Brown, 1996). This effect will be of great importance for the extension of the navigation season in the Russian Northern Sea Route and the NorthWest Passage.

Predicting the future character of moving pack ice is a difficult problem because dynamics (ocean and wind currents), rather than thermodynamics (radiation and heat components), determine its average thickness. Wind stress acting on the ice surface causes the ice cover to open up to form leads (ice-free areas). Later, under convergent stress, refrozen leads and thinner ice elements are crushed to form pressure ridges. Exchanges of heat, salt, and momentum are all different from those that would occur in fast-ice cover. The area-averaged growth rate of ice is dominated (especially in autumn and early winter, when much lead and ridge creation take place) by the small fraction of the sea surface occupied by ice less than 1 m thick (IPCC 1996, WG II, Section 7.4.5).

In the coastal zones of the Arctic Ocean-such as off the Canadian Arctic Archipelago, where there is net convergence of currents-the mean ice thickness is very high (7 m or more) because of ridging. Here the mean thickness is determined by mechanical factors, largely the strength of the ice, and may not be as sensitive to global warming as in other regions (IPCC 1996, WG II, Section 7.4.5).

Feedback mechanisms in the Arctic often are strong and complex, and not all have been fully identified or quantified. Most are positive in nature. For example, as biomes migrate northward, taller plants will tend to lower albedo, especially where they protrude through the snow. This will lead to a further enhancement of warming. Similarly, decreases in the extent of snow and ice cover will lower albedo and act to warm the water and land. Increased precipitation, expected with many warming scenarios, could provide a possible negative feedback as some will contribute to increased snow thickness. As Arctic warming increases the open-water area, precipitation may further increase and cause thicker snow cover, including on sea ice. The growth rate of land-fast ice could be expected to decrease, as has been directly observed (Brown and Cot�, 1992). However, if snow thickness is increased to the point where not all of it is melted in summer, more protection may be conferred on the ice surface-which could lead to an increase, rather than a decrease, in equilibrium ice thickness (IPCC 1996, WG II, Section 7.4.5). This scenario also would increase the stability of the upper mixed layer of the ocean, leading to more sea ice production.