Sea ice is a predominant feature of the polar oceans. Its extent expands and
contracts markedly from winter to summer. Sea ice has a dramatic effect on the
physical characteristics of the ocean surface. The normal exchange of heat and
mass between the atmosphere and ocean is strongly modulated by sea ice, which
isolates the sea surface from the usual atmospheric forcing (Williams et al.,
1998a). In addition, sea ice affects albedo, the exchange of heat and moisture
with the atmosphere, and the habitats of marine life. Finally, sea ice plays
a significant role in the thermohaline circulation of the ocean. Changes in
air temperature expected with projected climate change are likely to alter the
sea-ice regime and hence have impacts on the foregoing mechanisms.
Warming is expected to cause a reduction in the area covered by sea ice, which in turn will allow increased absorption of solar radiation and a further increase in temperature. In some climate models, this sea ice-albedo feedback has led to amplification of projected warming at higher latitudes, more in the Arctic Ocean than in the Southern Ocean. At some point, with prolonged warming a transition to an Arctic Ocean that is ice-free in summerand perhaps even in the wintercould take place. The possibility of a transition to an ice-free Arctic Ocean that is irreversible also must be considered. Climate models predict a wide range of changes for polar sea ice by the year 2100 (see TAR WGI Chapter 1). Many of the earlier models treated sea ice very simplistically and were not coupled with the ocean, so their results are unlikely to be reliable. More recent climate models include sophisticated sea-ice routines that take into account the dynamics and thermodynamics that control sea-ice formation, transport, and melt (Everett and Fitzharris, 1998). These models, however, still are limited in their ability to reproduce detailed aspects of sea-ice distribution and timing.
Snow on sea ice controls most of the radiative exchange between the ocean and atmosphere, but its exact role in energy transfer is uncertain (Iacozza and Barber, 1999) and difficult to model because of a lack of adequate data (Hanesiak et al., 1999; Wu et al., 1999). A further complication is the timing of snowfall in relation to formation of sea ice (Barber and Nghiem, 1999). Accumulation of snow on sea ice also plays a significant role in sub-ice primary production and for habitats of marine mammals (e.g., polar bear and seals).
Whether the sea ice in the Arctic Ocean will shrink depends on changes in the
overall ice and salinity budget, the rate of sea-ice production, the rate of
melt, and advection of sea ice into and out of the Arctic Basin. The most important
exit route is through Fram Strait (Vinje et al., 1998). The mean annual export
of sea ice through Fram Strait was ~2,850 km3 for the period 1990-1996,
but there is high interannual variability caused by atmospheric forcing and,
to a lesser degree, ice thickness variations. Other important passages are the
northern Barents Sea and through the Canadian Arctic Archipelago (Rothrock et al., 1999). Analyses by Gordon and O'Farrell (1997), using a dynamic ice
routine with a transient coupled atmosphere-ocean climate model (Commonwealth
Scientific and Industrial Research OrganisationCSIRO), predict a 60% loss
in summer sea ice in the Arctic for a doubling of CO2. The summer
season, during which ice retreats far offshore, increases from 60 to 150 days.
The likely distance between northern coasts and Arctic pack ice will increase
from the current 150-200 to 500-800 km.
In a more recent study, there is good agreement between Arctic sea-ice trends and those simulated by control and transient integrations from the Geophysical Fluid Dynamics Laboratory (GFDL) and the Hadley Centre (see Figure 16-6). Although the Hadley Centre climate model underestimates sea-ice extent and thickness, the trends of the two models are similar. Both models predict continued decreases in sea-ice thickness and extent (Vinnikov et al., 1999), so that by 2050, sea-ice extent is reduced to about 80% of area it covered at the mid-20th century.
GCM simulations for Arctic sea ice predict that warming will cause a decrease
in maximum ice thickness of about 0.06 m per °C and an increase in open
water duration of about 7.5 days per °C (Flato and Brown, 1996). These projections
are somewhat lower than changes observed during the latter part of the 20th
century (see Section 16.1.3). Increased snowfall
initially causes a decrease in maximum thickness (and corresponding increase
in open-water duration), but beyond 4 mm per day (1.33 mm per day water equivalent),
formation of "slush ice" by surface flooding offsets the insulating effect of
snow and causes an increase in maximum thickness and a decrease in the duration
of open water.
Without sea ice, wave heights will increase, and the Arctic coast will be more exposed to severe weather events such as storm surges that cause increased coastal erosion, inundation, and threat to structures. Areas of ice-rich permafrost are particularly vulnerable to coastal erosion (Nairn et al., 1998; Wolfe et al., 1998). A portent of the future is severe coastal erosionas much as 40 m yr-1along the Siberian coast. Deposited organic material changes the entire biogeochemistry of the nearshore waters (Weller and Lange, 1999). More open water may lead to increased precipitation and further amelioration of temperatures. Projected losses in sea ice are likely to have considerable impacts on Arctic biology through the entire food chain, from algae to higher predators (e.g., polar bears and seals). Loss of sea ice also will affect indigenous peoples and their traditional ways of life. A more open ocean will favor increased shipping along high-latitude routes and could lead to faster and cheaper ship transport between eastern Asia, Europe, and eastern North Amercia.
Other reports in this collection