Improving the parameterization of cloud and precipitation in the ECMWF global numerical weather prediction model

Description

The current cloud and precipitation parameterization scheme in the ECMWF global numerical weather prediction model is based on Tiedtke (1993), parameterizing the sources and sinks of cloud condensate and cloud fraction due to the main generation and destruction processes from convection and microphysics. The cloud condensate is diagnostically separated into liquid and ice as a function of temperature, and rain and precipitating snow are treated diagnostically. Although the scheme has performed well in the ECMWF global model, it is clear that a number of important cloud microphysical processes cannot be well represented, particularly for mixed-phase clouds. Recent developments to the cloud scheme have introduced separate prognostic variables for liquid and ice, rain and snow with corresponding changes to the parameterization of microphysical processes. Preliminary evaluation of the new scheme shows an improved representation of mixed-phase clouds, ice water path, and the advection of snow. The potential for further evaluation with remote sensing data from CloudSat/CALIPSO and the ARM site facilities is highlighted.