The prediction of cloud droplet number concentration for both cloud droplet modes has been implemented through the application of a Lagrangian parcel model. This model takes a user specified CCN spectrum and predicts on the number and mass of CCN activated. Parcel model results are included in the mesoscale model thru the use of lookup tables that vary with vertical velocity, temperature, concentration of CCN, and median radius of the nuclei spectrum. CCN in the parcel model are allowed to reach supersaturation and maximum growth for inclusion in the lookup tables.
The parcel model and resultant lookup tables take the user specified number concentration of CCN and provide a percentage of CCN nucleated that depends upon the above mentioned atmospheric conditions. The RAMS model then retrieves the percentage nucleated from the tables and calculates the number and mass of CCN to activate and nucleate into cloud droplets. The general microphysics then provides sources and sinks for the CCN, and it keeps track of the transition of CCN mass contained within the hydrometeors as they change phase, size, and categories; this allows for more accurate restoration of CCN mass and number in the case of evaporation which returns CCN into the atmosphere.
RAMS has also been fitted with the option to specify and nucleate giant-CCN (GCCN) with median radius from 1-5 microns. Results of multiple parcel model simulations reveal that nearly all GCCN nucleate in most situations. With the average number concentration of GCCN being several orders of magnitude smaller than that of CCN, the CCN are dominant in the nucleation process. Examination of parcel model results, using a dual-mode GCCN/CCN spectrum, revealed that the presence of GCCN did not significantly alter the percentage of CCN that are allowed to activate. From this result it was decided to allow the user to specify the number concentration of GCCN and then nucleate 100% of these prior to nucleation of CCN. Given the small numbers of GCCN present, their nucleation occurs rather quickly and then allows excess vapor to be used to nucleate smaller CCN.
Use of two modes of cloud droplets better predicts rain formation by the introduction of a large droplet (drizzle) mode with diameters from 40-80 microns in addition to the small droplet mode (< 40 microns). This addition allows the nucleation of CCN into the small droplet mode and the GCCN into the large droplet mode. The dual-mode also has the effect of slowing the production of rain by forcing droplet growth, by vapor deposition and collision/coalescence, to pass through the observed two modes of cloud droplets before entering the rain category. In addition, the presence of two modes provides a more realistic simulation of the Hallett-Mossop secondary ice formation process.
Initial results support observations such that increased numbers of GCCN in combination with fewer CCN reduces the amount of cloud water and increases the amount of rain and accumulated precipitation. Further results of simulations with varying amounts of CCN and GCCN in situations involving deep convection, stratiform precipitation, mixed-phase clouds and cold-weather icing conditions will be presented.
Supplementary URL: