Prognostic Evaluation of Assumptions Used by Cumulus Parameterizations

Abstract

Using a spectral-type cumulus parameterization that includes moist downdrafts within a three-dimensional mesoscale model, various disparate closure assumptions are systematically tested within the generalized framework of dynamic control, static control, and feedback. Only one assumption at a time is changed and tested using a midlatitude environment of severe convection. A control run is presented, which shows good agreement with observations in many aspects. Results of the sensitivity tests are compared to observations in terms of sea level pressure, rainfall patterns, and domain-averaged bias errors (compared to the control run) of various properties. The dynamic control is the part that determines the modulation of the convection by the environment. It is shown that rate of destabilization, as well as instantaneous stability, work well for the dynamic control. Integrated moisture convergence leads to underprediction of rainfall rates and subsequent degrading of the results in terms of movement and structure of the mesoscale convective system (MCS). The feedback determines the modification of the environment by the convection, and in this study is considered together with the static control, which determines cloud properties. All feedback and static-control assumptions tested here seem very important for the prediction of sea level pressure and rainfall. The most crucial ones were downdrafts and lateral mixing. As an interesting by-product, it is shown that a very simplistic and computationally highly efficient convective parameterization scheme leads to a very realistic simulation of the MCS, if the scheme uses a stability closure, assumes a large cloud size, parameterizes moist downdrafts, and does not assume unrealistically law lateral mixing.