On using mixed-layer transport parameterizations with radiometric surface temperature for computing regional scale sensible heat flux

Recent mixed-layer formulations for computing large-scale surface energy fluxes under daytime convective conditions do not require the estimation of surface-layer parameters, such as the roughness lengths for momentum and heat. This greatly simplifies approaches using operational satellite measurements of surface temperature for computing the surface energy balance at regional scales because the surface roughness parameters are not well known for many landscapes. The utility of such mixed-layer formulations is tested using data from several recent multidisciplinary field experiments (HAPEX-MOBILHY, FIFE and Monsoon 90). The results indicate that specific mixed-layer formulations adequately simulate surface sensible heat fluxes in the grassland and shrubland sites. However, use of the original values of proposed empirical coefficients for the forested site yield poor results. This is probably due to the fact that the forested site has significantly different surface geometry and associated distribution of temperature among the surface components (especially the relative importance of soil background temperatures) compared to the other sites. Therefore, the relationship between aerodynamic and radiometric surface temperature may have greatly differed between the forested site and the other locations. However, differences in aerodynamic roughness between the experimental sites were not correlated with changes required in the values of the coefficients. Instead, a two-source model which makes the distinction between aerodynamic and radiative temperature is proposed, as a means to determine which surface properties significantly affect the magnitude of the mixed-layer coefficients.