Forecasts and simulations are varied owing to different allocation of 3-dimensional variables in mesoscale models. No attempts have been made to address the issue of optimizing the simulation with a 3-dimensional vari...Forecasts and simulations are varied owing to different allocation of 3-dimensional variables in mesoscale models. No attempts have been made to address the issue of optimizing the simulation with a 3-dimensional variables distribution that should come to be. On the basis of linear nonhydrostatic anelastic equations, the paper hereby compares, mainly graphically, the computational dispersion with analytical solutions for four kinds of 3-dimensional meshes commonly found in mesoscale models, in terms of frequency, horizontal and vertical group velocities. The result indicates that the 3-D mesh C/CP has the best computational dispersion, followed by Z/LZ and Z/LY, with the C/L having the worst performance. It is then known that the C/CP mesh is the most desirable allocation in the design of nonhydrostatic baroclinic models. The mesh has, however, larger errors when dealing with shorter horizontal wavelengths. For the simulation of smaller horizontal scales, the horizontal grid intervals have to be shortened to reduce the errors. Additionally, in view of the dominant use of C/CP mesh in finite-difference models, it should be used in conjunction with the Z/LZ or Z/LY mesh if variables are allocated in spectral models.展开更多
基金Supported by the open research program of LASG Institute of Atmospheric Physics Chinese Academy of Sciences
文摘Forecasts and simulations are varied owing to different allocation of 3-dimensional variables in mesoscale models. No attempts have been made to address the issue of optimizing the simulation with a 3-dimensional variables distribution that should come to be. On the basis of linear nonhydrostatic anelastic equations, the paper hereby compares, mainly graphically, the computational dispersion with analytical solutions for four kinds of 3-dimensional meshes commonly found in mesoscale models, in terms of frequency, horizontal and vertical group velocities. The result indicates that the 3-D mesh C/CP has the best computational dispersion, followed by Z/LZ and Z/LY, with the C/L having the worst performance. It is then known that the C/CP mesh is the most desirable allocation in the design of nonhydrostatic baroclinic models. The mesh has, however, larger errors when dealing with shorter horizontal wavelengths. For the simulation of smaller horizontal scales, the horizontal grid intervals have to be shortened to reduce the errors. Additionally, in view of the dominant use of C/CP mesh in finite-difference models, it should be used in conjunction with the Z/LZ or Z/LY mesh if variables are allocated in spectral models.
