Climate simulations with a new air-sea turbulent flux parameterization in the National Center for Atmospheric Research Community Atmosphere Model (CAM3)

This study examines climate simulations with the National Center for Atmospheric Research Community Atmosphere Model version 3 (NCAR CAM3) using a new air-sea turbulent flux parameterization scheme. The current air-sea turbulent flux scheme in CAM3 consists of three basic bulk flux equations that are solved simultaneously by an iterative computational technique. We recently developed a new turbulent flux parameterization scheme where the Obukhov stability length is parameterized directly by using a bulk Richardson number, an aerodynamic roughness length, and a heat roughness length. Its advantages are that it (1) avoids the iterative process and thus increases the computational efficiency, (2) takes account of the difference between z₀m and z₀h and allows large z₀m/z₀h, and (3) preserves the accuracy of iteration. An offline test using Tropical Ocean–Global Atmosphere Coupled Ocean-Atmosphere Response Experiment (TOGA COARE) data shows that the original scheme overestimates the surface fluxes under very weak winds but the new scheme gives better results. Under identical initial and boundary conditions, the original CAM3 and CAM3 coupled with the new turbulent flux scheme are used to simulate the global distribution of air-sea surface turbulent fluxes, and precipitation. Comparisons of model outputs against the European Remote Sensing Satellites (ERS), the Objectively Analyzed air-sea Fluxes (OAFlux), and Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) show that: (1) the new scheme produces more realistic surface wind stress in the North Pacific and North Atlantic trade wind belts and wintertime extratropical storm track regions; (2) the latent heat flux in the Northern Hemisphere trade wind zones shows modest improvement in the new scheme, and the latent heat flux bias in the western boundary current region of the Gulf Stream is reduced; and (3) the simulated precipitation in the new scheme is closer to observation in the Asian monsoon region.