Application of Transient Turbulence Theory to Mesoscale Numerical Weather Forecasting

摘要

A transient turbulence parameterization has been inserted into Penn State/NCAR regional three dimensional mesoscale numerical model. Transient turbulence parameterizes both boundary layer and clear air turbulence (CAT). The new transient turbulence scheme replaces separate formulations of the boundary layer parameterization, dry convective adjustment, cumulus parameterization and horizontal diffusion. A new surface layer flux formulation that is compatible with the transient turbulence parameterization has also been installed. Additionally, a sixth order implicit tangent filter has been inserted to remove numerical noise. This separation of the turbulence physics from numerical stability considerations gives us an opportunity to examine the mixing process in greater detail. We compared forecasts from our revised model with the standard Penn State/NCAR model formulation, which has a Blackadar boundary layer and fourth order K theory horizontal diffusion. Forecasts using both Kuo cumulus parameterization and explicit cloud convective representations have been made with this standard model utilizing the OSCAR IV and CAPTEX data sets. Comparisons between the control and the transient turbulence approach are used to highlight the role of turbulence and its affect on the overall forecast. Verifying station radiosonde reports and analysis are the final authority on the accuracy of the forecasts. Keywords: Clear air turbulence, Surface fluxes, Weather prediction. (KR)