Sea Ice Dynamics and Thermodynamics
Programme Background
Results from GCM climate change experiments indicate that the largest warming is expected over the Arctic, ice-covered ocean. The expected decline in the fresh, 2-3000 km3 yr-1 sea ice flux from the Arctic to the Greenland-Iceland-Norwegian (GIN) Seas may offset the effect on thermohaline circulation of the increase in precipitation or ice sheet run-off in a warming atmosphere, estimated from GCM experiments as ~ 700 km3 yr-1 K-1. Submarine data suggests that a decrease of 1.3 m in the mean draft of Arctic sea ice has occurred since 1958 to 1995. The sensitivity of the thermohaline circulation to these competing processes makes the detailed treatment of sea ice in GCMs important. GCM improvements, particularly the removal of flux corrections, have increased confidence in their predictions concerning sea ice. Model predictions have been used to argue that an observed decrease in sea ice extent is anthropogenic in origin. However, of the leading climate models, HADCM2 underestimated Arctic summer ice area by 50%, whereas HADCM3 overestimates winter ice area by 25%. While regional Arctic sea ice models are able, with forecast wind fields, to reproduce quite well fluctuations in ice flux through the Fram Strait, this is achieved through the ad-hoc tuning of parameter values such as ice strength, the drag coefficients of ocean- and wind-induced stress. New satellite data indicates that even such state-of-the-art regional models can be in error by a factor of 2 in ice thickness, and hence mass.
There has been a fundamental lack of knowledge regarding the regional, seasonal and interannual variability of Arctic sea ice thickness. This lack of knowledge has limited improving the treatment of sea ice in regional models, most of which are based on Hibler s 1979 dynamic parameterisation. The central objectives of this programme are: " To bring new validated satellite observations to bear on the testing of existing GCM and regional sea ice models, particularly of the Arctic, and " Develop new models of sea ice whose description of the physics governing the evolution of the ice pack will be a generation ahead of existing regional models.
Return to Research Programmes