Abstract
Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) are used to assess trade cumulus cloudiness in three versions of the Integrated Forecasting System of the European Centre for Medium-Range Weather Forecasts. The observations are recast onto the model grid, and two simple threshold criteria for cloud-top height and cloud fraction are used to identify grid points containing trade cumulus clouds. The cloud fraction and cloud-top height distributions of the sample populations are then compared. Results show that all versions of the model overestimate the frequency of occurrence of trade cumulus clouds but underestimate their cloud fraction when present. These effects partially compensate. Cloud-top heights are overestimated in model cycles using the modified Tiedtke parameterization for shallow convection, but are in very good agreement with observations when the dual mass flux parameterization is introduced.
Corresponding author address: Maike Ahlgrimm, European Centre for Medium-Range Weather Forecasts, Shinfield Park, Reading RG2 9AX, United Kingdom. Email: maike.ahlgrimm@ecmwf.int