Premium
The sensitivity of simulated orographic precipitation to model components other than cloud microphysics
Author(s) -
Zängl G.
Publication year - 2004
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1256/qj.03.119
Subject(s) - parametrization (atmospheric modeling) , orographic lift , mesoscale meteorology , mm5 , orography , precipitation , meteorology , sensitivity (control systems) , boundary layer , discretization , cloud physics , physics , environmental science , mechanics , cloud computing , computer science , mathematics , radiative transfer , mathematical analysis , quantum mechanics , electronic engineering , engineering , operating system
Numerical simulations of an orographic precipitation event observed during the MAP field phase are presented in order to investigate side effects on the simulated precipitation. The simulations are conducted with the Penn State University—National Center for Atmospheric Research mesoscale model MM5 in a configuration with four interactively nested domains and a finest grid spacing of 1.4 km. The sensitivity tests examine the impact of the boundary‐layer parametrization, the set‐up of the convection parametrizations in the outer model domains, the implementation of the horizontal diffusion and the specification of the vertical coordinate. For reference, five different cloud microphysical parametrizations are compared against each other. The results show that each of the side effects arising from the model numerics and from the physics parametrizations not related to cloud microphysics can be as large as, or even larger than, the impact of the microphysical parametrization itself. A particularly large sensitivity is found for the horizontal diffusion of temperature and moisture. Computing this diffusion along the terrain‐following coordinate surfaces rather than truly horizontally increases the average precipitation in the 1.4 km domain by about 35%, and a comparison with surface observations indicates a pronounced deterioration of the model results. The other sensitivities are found to range between 10% and 15% on a domain average. An important conclusion emanating from these results is that future model development should pay much more attention to the cross‐sensitivities between the model numerics, the convection and boundary‐layer parametrizations and the cloud microphysical parametrization. Copyright © 2004 Royal Meteorological Society.