摘要: We reviewed cloud physics parameterizations comprehensively in order to examine their theoretical bases and to evaluate their applicability to mesoscale modeling. We found that the constant distribution function of hydrometeors as... 展开 We reviewed cloud physics parameterizations comprehensively in order to examine their theoretical bases and to evaluate their applicability to mesoscale modeling. We found that the constant distribution function of hydrometeors assumed in previous work are unrealistic. To circumvent this problem, we have made new parameterizations by multiple regression on mixing ratios of water vapor, cloud water, rainwater, cloud ice, graupel, and snowflakes. We used solution fields derived from simulations with a cloud model that incorporates sophisticated treatments of microphysics and dynamics. In the detailed simulations from which they were derived, the regression parameterizations generally give better results than do the existing parameterizations tested. In a Lagrangian air parcel model under variable atmospheric conditions, the new parameterizations give consistent results and are computationally more efficient than previously available parameterizations. 77 refs., 35 figs., 4 tabs. 收起