Impact of a two-moment cloud model on the microphysical structure of two precipitation events

Abstract

A new two-moment microphysical parameterization is described. The proposed scheme predicts the mixing ratio and number concentration of rain, pristine ice crystals, snow, aggregates, graupel and hail. The general gamma distribution is the basis function used for each hydrometeor. Preliminary sensitivity testing of the new microphysical scheme in an idealized convective simulation shows that the two-moment prediction scheme allows more freedom on the size distribution enabling the diameter to evolve more realistically than in a 1-moment scheme. Sensitivity to the prescribed input parameters such as cloud droplet concentrations and the shape parameter v is demonstrated in the model results. Model verification is performed on two separate case studies which occurred in very diverse environments. A wintertime shallow post-frontal upslope case from the Winter Icing and Storm Project (WISP91) and a strong summertime convective storm from Cooperative Convective Precipitation Experiment (CCOPE81) are investigated. In these case studies, the kinematic evolution of the two-moment and one-moment simulations is very similar, however, the microphysical structure shows differences between the two schemes. In the two-moment simulation, the diameters of each hydrometeor are allowed to evolve depending on the environmental conditions, and these fields compare well to the observed measurements. In the one-moment simulations, with either the mean diameter or the slope intercept parameter fixed, the results show the unrealistic constraint of not allowing variations in the hydrometeor spectra. Comparisons to aircraft measurements for both cases show that the microphysical structure is predicted well by the two-moment scheme compared to the observations. Peak reflectivities are predicted close to the observed values using the two-moment scheme, while the one-moment scheme with the mean diameter specified over-predicts reflectivities due to the larger prescribed diameters. In the simulation where No is specified, reflectivity are similar to the observations, however, other microphysical parameters such as graupel and rain are not simulated well. In both one moment simulations, the ratio of hail precipitation to rain is grossly over-predicted. Sensitivity to variations in the v parameter are demonstrated in the summertime case. These sensitivity experiments show the need for more measurements of the size and shape of the hydrometeor spectra for various weather environments. There is also a need to numerically examine the effects of varying 11 in a variety of environments and storms.