Evaluation of GEM in the Stable Boundary Layer over the Arctic Ocean during SHEBA

Climate change amplification projected by IPCC models over the Arctic Ocean is very uncertain because Global Climate Models perform poorly for representing the Atmosphere-Ice-Ocean Interaction that takes place in the atmospheric boundary layer. Vertical mixing and turbulent fluxes of heat, moisture and momentum are reduced efficiently by the deep temperature inversion that develops in the stable boundary layer.

Tjernström et al. (2004) indicate that models participating in the ARCMIP project (Arctic Regional Climate Model Intercomparison Project) overestimate surface fluxes of heat and moisture over sea-ice in comparison with SHEBA (Surface Heat Budget of the Arctic Ocean) observations. Errors up to 200% are reported in ARCMIP simulations for near-surface winds in weak wind conditions. Moreover, Grachev et al. (2005) have shown that the Busigner-Dyer profiles used to parameterize vertical profiles of wind velocity, temperature and humidity in atmospheric models perform poorly in very stables conditions over the Arctic Ocean.

The goal of this project is to evaluate the Canadian Global Environmental Multiscale (GEM) Model to simulate the Arctic climate near the surface during the SHEBA year. Also, an evaluation is made of GEM's performances to simulate surface fluxes in function of stability and to verify if performances are decreasing as stability increases.