Numerical simulation of a mesoscale convective complex: Model development and numerical results

A mesoscale numerical model has been developed and used to study the environment which supported the development of a mesoscale convective complex (MCC). The hydrostatic numerical model was combined with the CSU cloud/mesoscale model in 1983 to form the CSU Regional Atmospheric Modelling System (RAMS). Some of the aspects of RAMS developed were a hydrostatic "time-split" time differencing scheme, a soil model, a higher-ordered forward upstream advection scheme, improved versions of two convective parameterization schemes, and an isentropic data analysis package.; The goal of the numerical simulations was to employ the numerical model to study an MCC with higher space and time resolution than is available through observational means. While there were many differences, the coarse resolution control run simulated an MCC whose meso-{dollar}alpha{dollar}-scale structure evolved similarly with the observed convective system to establish the credibility of the model.; Higher resolution simulations, made to increase the spatial resolution, compared favorably with the coarse resolution simulations. These results were examined for the important forcing mechanisms of this MCC. An important mechanism was the development and propagation of the mountain/plains solenoidal circulation which was forced by the baroclinicity created by the physiographic features of the topography slope and horizontal gradients of soil moisture. Other factors that were hypothesized to be important were low-level "heat low", the Bermuda high, a weak front moving southward from Canada, and an upper level jet core.; Results from a simplified two-dimensional simulation verified many features of the solenoid's behavior. The solenoid may also be responsible for the nocturnal preference for MCCs and the frequently observed mid-level shortwave that often accompanies the convective systems.; Two higher resolution sensitivity simulations were performed. The first, in which the convective parameterization was not used, produced a low-level solenoidal circulation which, by the end of the simulation, looked very similar to the solenoid in the control run. The second sensitivity experiment with the resolved microphysical parameterizations activated showed that the gross behavior of the MCC was similar to the control run although there were differences in the details of the mesoscale vertical motion fields and convection.