Effects of physics packages on medium-range forecasts in a global forecasting system

The skill of medium-range forecasting can be determined by the quality of the initial conditions through data assimilation, model dynamics, and physics. In this study, the roles of physics packages in global forecasting systems are evaluated on a medium-range forecast testbed. Two physics packages are evaluated: The standard physics package in the model is a version of the National Centers for Environmental Prediction (NCEP) global forecast system, which has been employed to produce the NCEP/Department of Energy (DOE) reanalysis. The revised package consists of schemes accounting for radiation, the planetary-boundary layer, the land-surface model, gravity wave drag, and moist convection. Evaluations were performed in August 2010 and January 2011, focusing on the 500-hPa geopotential height and precipitation. The results show that the revised physics package improved the performance of the 500-hPa geopotential height throughout the 10-day forecast period. The improvement is larger in the southern hemisphere than in the northern hemisphere after the 5-day forecast, with a more distinct enhancement in the winter hemisphere. The skill of the precipitation forecasts shows overall improvement, although too much suppression of heavy precipitation needs to be rectified. The cumulus parameterization scheme is found to be a critical factor in improving the large-scale flow, but revisions in other processes, such as the mutual compensation effects among physics algorithms, need to be considered. (c) 2013 Elsevier Ltd. All rights reserved.