Sea ice climatology, variations and teleconnections: Observational and modeling studies.

摘要

Hypotheses, models and observations suggest that sea ice plays an important role in the local, regional and global climate through a variety of processes across a full range of scales. However, our documentation and understanding of the nature of the polar-extrapolar climate teleconnections and their underlying causal and mechanistic links are still rudimentary, and the largest disagreements among model simulations of present and future climate are in the polar regions.; In an effort to address these issues, we evaluated the simulated Antarctic sea ice variability and its climate teleconnections in three coupled global climate models (GISS, NCAR and GFDL) as compared to the observations. All the models capture the El Niño-Southern Oscillation (ENSO)-like phenomenon to some degree, although almost all the models miss some observed linkages. The GISS and NCAR models also capture the observed Antarctic Dipole and meridional banding structure through the Pacific. The Antarctic sea ice regions showing the strongest sensitivity to global teleconnections differ among the models and between the models and observations. We then proposed that the changes of the regional mean meridional atmospheric circulation (the regional Ferrel Cell) are one such mechanism leading to the covariability of the ENSO and Antarctic Dipole by modulating the mean meridional heat flux using the observational data.; To more accurately represent sea ice simulations and associated feedbacks with the atmosphere and the ocean, the GISS coupled model was used to investigate the sensitivity of sea ice to the following physical parameterizations: (a) two sea ice dynamics (cavitating fluid and viscous-plastic), (b) the specification of oceanic isopyncal mixing coefficients in the Gent and McWillams isopyncal mixing, (c) the Wajsowicz viscosity diffusion, (d) surface albedo, (e) the penetration of solar radiation in sea ice, (f) effects of including a sea ice salinity budget, and (g) the ice-ocean boundary interactions. The atmospheric responses associated with sea ice changes were discussed. Based on these experiments, a series of composite experiments with the aforementioned parameterizations were also investigated. While improvements are seen overall, there are some unrealistic aspects that will require further improvements to the sea ice and ocean components. Finally, the GISS coupled model's ability to represent observed Antarctic sea ice variability and its global teleconnections was re-evaluated in a composite run with the improved sea ice and ocean processes.