Quantitative precipitation forecasts in the warm season have historically been much poorer than in the cool season. To a large degree, this is because a sizeable fraction of warm season precipitation comes from mesoscale convective systems (MCSs) and these systems are poorly resolved by the current operational models. Recent observational and budget studies indicate that numerical prediction of these systems will require physical processes such as precipitation drag, melting and evaporation to be incorporated into very fine resolution mesoscale models. The convective parameterization scheme used with such models must include moist downdrafts and realistic moisture detrainment for propagating deep convection. Part of the proposed work focuses on the model development necessary to successfully simulate MCSs. A second portion focuses on the use of the model as a diagnostic tool to help understand the structure and dynamics of MCSs, particularly the development of warm core vortices in the "stratiform" region. The final portion concentrates on observing-system simulation experiments in order to help define the observational requirements (including assimilation of profiler data) for understanding and predicting mesoscale convective systems.
