A Coupled Atmosphere - Sea Ice Model Study of the Role of Sea Ice in Climatic Predictability

• Main Author: Bennett, Theodore Joseph
• Format: Text
• Language: unknown
• Published: 1982
• Subjects: Physics; Atmospheric Science; Sea ice
• Online Access: http://hdl.handle.net/2142/77341

219 p. Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1982. Two ensembles of four October 1 to December 10 simulations with a coupled atmosphere-sea ice model are made and a 30-day mean climatic state is calculated for each ensemble for the period of November 11 to December 10. Initially, the light ice ensemble has a thick ice cover extending southward to 75(DEGREES)N. The heavy ice ensemble has, in addition, a 0.25 m thick layer of ice covering 75% of the area along latitudes 69(DEGREES)N and 64(DEGREES)N. The southward advance of sea ice is limited by the model to 64(DEGREES)N. The two ensemble-mean climatic states are compared to determine how the initial difference in ice cover evolves with time and whether statistically significant differences in the atmospheric circulation exist. The atmosphere is described by a hemispheric, two-layer model based on the spectral form of the linear balance system of equations. The physical processes which are included in the model are longwave and shortwave radiation, the exchange of sensible and latent heat at the earth's surface, convective adjustment, friction, diffusion, and, in simple form, latent heat release due to precipitation. The thickness and temperature of the ice and snow, the fraction of open water, and the temperature of the polar oceanic mixed layer are modeled. The results of this experiment include (1) the differences between the ensemble-mean ice covers are principally differences in the thickness of ice and snow and fraction of open water rather than a difference in ice extent. These differences have considerable zonal asymmetry. (2) There are also regional differences in the 30-day mean potential temperature, stream function, and zonal wind at 750 mb and surface temperature which are statistically significant at 10% to 1% or better. (3) It is difficult to give a cause and effect description of the relationship between the differences in the time mean 750 mb stream function and spectrally filtered heating field. (4) The flux of sensible heat, area-weighted over ice and leads, is found to depend strongly on the thickness of ice and snow and fraction of open water. (5) The results of a series of sensitivity tests of the sea ice model are reported.