Application of An Immersed Boundary Method to Variational Doppler Radar Analysis System

The Variational Doppler Radar Analysis System (VDRAS) has been widely used for nowcasting and even quantitative precipitation forecast (QPE). Although many successful applications done by VDRAS were already shown recently, the lack of ability to resolve terrain effect degrades its performance when encounter the environment with complex topography. In order to improve the analysis system, Ghost Cell Immersed Boundary Method (GCIBM) is applied to the system's cloud model. The GCIBM can be implemented in any Cartesian grid code, and the boundary conditions are enforced through a ghost cell grid point. Since the coordinate of VDRAS cloud model is also Cartesian, GCIBM is convenient for the implementation. In this study, the validation experiment is accomplished by a two-dimensional linear mountain wave simulation, and the result is compatible with analytical solution. Furthermore, a three-dimensional leeside vortex simulation is also conducted, the extension of a pair of symmetric vortices represents good agreement with previous researches. Last, a real case forecast experiment is initialized by a VDRAS 4DVAR analysis field which assimilates radar data but doesn't have ability of terrain resolving. The terrain effect is obviously shown by dynamic and thermodynamic variables after forecasting by new forward model, which results improvement on QPE performance at the same time.