Atmospheric boundary layer modeling in complex terrain

The atmospheric boundary layer (ABL) modeling in complex terrain is one of the most important parts not only in basic atmospheric sciences, but also in applied meteorology. The parameterization of surface fluxes in heterogeneous area is one of the priorities in ABL research.; In our research work, a new approach to estimate the surface fluxes in heterogeneous area is proposed using the "blending height" concept. The blending height is a scale height such as that the flow above it within the surface layer does not depend on the heterogeneous surface features. In this approach, the blending height is derived from the neutral condition. The meteorological variables at the blending height are obtained by using the effective roughness lengths and the effective surface temperature. Then the surface fluxes are estimated by areally-averaged method from the local ones which are estimated under the blending height. By this new approach, the fluxes can meet well with that estimated from the effective parameters above the blending height. The results are satisfactory by comparison with that from the theoretical model.; Meanwhile, the original Louis scheme is improved by relaxing its two assumptions. The improved scheme is suitable for different surfaces and a wide range of atmospheric stability. It shows smaller errors by comparison with the theoretical results. This scheme can be coupled with the new approach for estimation of surface fluxes to save computational time. They can be applied for the atmospheric modeling system.; The Regional Atmospheric Modeling System (RAMS) is improved by the new approach and the improved Louis scheme for estimation of surface fluxes. The model evaluation shows the improvements are very important and necessary. The improved RAMS can be applied to do the ABL modeling in the Pearl River Delta, Southern China, having a very complex terrain.; The numerical experiment is done based on considering the objectives and selecting the RAMS configuration. The modeling results show good agreement with the observations from both sounding and surface stations. The sensitivity tests on some parameters, initialization and boundary conditions are done to show the appropriate configuration for ABL modeling.