A New Non-Iterative Scheme for Surface Fluxes Parameterization

Abstract:

In weather or climate models, the earth’s surface is the boundary that needs to be resolved physically. The condition of atmosphere aloft (e.g., wind, temperature and humidity) is highly dependent on the momentum, sensible heat and latent heat fluxes at surface. However, parameterization of surface turbulent fluxes under unstably/stably stratified conditions has always been a challenge. Currently, the exchanges of momentum and heat fluxes between the earth’s surface and the atmosphere are usually calculated with various schemes based on Monin–Obukhov similarity theory. These schemes either need iterations or suffer low accuracy, which might consume excessive CPU time or could lead to unrealistic simulation results. In this paper, a non-iterative scheme is proposed to approach the classic iterative computation results using multiple regressions.The range −5 ≤ Ri_B ≤ 2.5, 10 ≤ z/z_0m ≤ 10⁵ and −0.5 ≤ ln(z_0m/z_0h) ≤ 30 is divided into several regions, and in each of the regions, multiple linear regression is performed to obtain non-iterative solutions for surface fluxes. As compared to the other most recent non-iterative schemes, we show that the suggested scheme has the smallest bias. The maximum relative errors of turbulent transfer coefficients for momentum (C_M) and sensible heat (C_H), as compared to those obtained from the classic iterative method, are always smaller than 2% in unstable condition and 12% in stable condition from our new non-iterative scheme.