| The evolution of clouds and the interaction with general circulation and the earth surface play an extremely important role in weather-climate systems. Microphysical schemes of atmospheric meso-scale numerical models directly predict behaviors of all cloud physical variables and their interrelations. When the schemes describe the microphysical processes in a more exact way, the techniques for precipitation predicting and the reliability of research results of the mechanism of generation and development for disastrous weather using the high resolution model results should be improved. As the higher model spatial resolution, the work of developing and improving microphysical schemes receive more attention. Due to limited observational data of cloud physical characteristics, the ability of schemes in reproducing the microphysical processes could hardly be directly proved in the work of predecessors. The deeper causes of diversity among the macro-fields simulated by various schemes are untraceable. The double-moment Bulk Microphysics Schemes have a reasonable physical background avoiding the deficiencies of the too-simple technologies in single-moment schemes and the huge computation cost of explicit bin Schemes, have achieved good simulation results and been widely used.In this paper, two double-Moment Bulk Microphysics Schemes, Milbrandt2-mon(MY)and Morrison2-mon(MOR), were compared using the WRF single-column model during the period of the Tropical Warm Pool International Cloud Experiment (TWP-ICE) field experiment. The conclusions could offer valuable references for schemes development.Results from the control simulations with the default settings of the two microphysics schemes in the WRF were able to reasonably reproduce the characteristics of the rain rate, the radiation fields, the liquid water content, and the frozen water content, as compared with observations. Significant deviations mainly occur during the monsoon suppressed period when the shallow convection perform too active and the ice-phase clouds are thicker compared with observations for both scnemes. The budgets of ice crystals and snow particles mach well with the distribution characteristics of ice-phase clouds. During the active monsoon period, the melting of ice-phase particles is one of the most important factors for precipitation formation, while the meilting of ice crystals in the MOR scheme only makes indirect contribution to the raindrops developments.The composition of the ice-phase clouds show different features between two schemes: ice-phase clouds are mainy composed of ice crystals and snow particles. Ice crystals in MY scheme extends to higher levels than those in MOR scheme; snow particles in MOR scheme are developed more strongly and make more contributions to ice clouds in the upper air than in MY scheme. Sensitivity tests about the assumption of constant snow bulk density show that different treatments in snow bulk density may cause the composition differences in the frozen water. The constant snow bulk density in MOR scheme is set too small, favoring the deposition growth of snow crystals and resulting in consuming much water vapor but suppressing ice crystals growth. In contrast, MY scheme assumes SBD varying inversely with diameter of snow crystal in MY scheme, supported by observations, resulting in less active snow growing together with ice crystals more advantageous in ice clouds.In addition, sensitivity tests about the total number of ice nuclei in deposition mode and condensation freezing mode (NIN) suggested that the responses of OLR and macro and micro physical characteristics of ice-phase clouds to NIN are linear during the depressed monsoon period, while active period show irregular responses to NIN.For MY scheme, the situation is complicated during all the period, where the development tendency of microphysical processes are concerned with specific weather conditions. The great response differences between two schemes are possibly related to the parameterization of primary nucleation of ice crystals. |
PDF Full Text Request