Key points are not available for this paper at this time.
The Pennsylvania State University/NCAR mesoscale model, originally developed by Anthes and Warner, is modified to simulate the meso-β scale structure and evolution of convectively driven weather systems. The modifications include: (i) two-way interactive nested-grid procedures, (ii) the Fritsch-Chappell convective parameterization scheme, and (iii) the Blackadar boundary layer package. An 18-h simulation of the Johnstown flood of July 1977 is conducted. Compared to the documentation of Hoxit et al. and Bosart and Sanders, the simulation reproduced many of the different aspect of the mesoscale convective complex and squall line that were responsible for the heavy rain over western Pennsylvania. In particular, the model predicts the size, propagation rate and orientation of the mesoscale convective components that were observed in the mid-Atlantic states. The simulated evolution of the planetary boundary layer, cool outflow boundaries and surface pressure perturbations, such as meso-β scale lows, highs, ridges and troughs, compare favorably to observations. Other mesoscale features, for example, low-level jets and maximum/minimum horizontal wind couplets associated with the convective systems, were also reproduced reasonably well. Of particular significance is that the model-forecast rainfall amounts and distribution are similar to the observed. Recognizing that a single case study does not provide a rigorous test of the predictability of a model, the results suggest that it may be possible to forecast the meso-β scale structure and evolution of convective weather systems with useful skill for periods up to about 18 hours. The results also suggest that significant improvements in warm-season quantitative precipitation forecasts might be possible if numerical forecasts of the meso-β scale structure and evolution of convective weather systems became operational.
Zhang et al. (Mon,) studied this question.