Parameterization of cloud droplet size distributions : 1 comparison with parcel models and observations

This work examines the efficacy of various physically-based 3 approaches derived from 1-D adiabatic parcel model frameworks (a numer-4 ical model and a simplified parameterization) to parameterize the cloud droplet 5 distribution characteristics for computing cloud effective radius and auto-conversion rate in regional/global atmospheric models. Evaluations are car-7 ried out for integrations with single (average) and distributions of updraft 8 velocity, assuming that a) conditions at s max are reflective of the cloud col-9 umn, or, b) cloud properties vary vertically, in agreement with 1-D parcel 10 theory. The predicted droplet distributions are then compared against in-11 situ cloud droplet observations obtained during the CRYSTAL-FACE and 12 CSTRIPE missions. Good agreement of droplet relative dispersion between 13 parcel model frameworks indicates the parameterized parcel model essentially 14 captures 1-D dynamics; the predicted distributions are overly narrow, with 15 relative dispersion being a factor of 2 lower than observations. However, if 16 conditions at cloud maximum supersaturation are used to predict relative 17 dispersion and applied throughout the cloud column, better agreement is seen 18 with observations, especially if integrations are carried out over the distri-19 bution of updraft velocity. When considering the efficiency of the method, 20 calculating cloud droplet spectral dispersion at s max is preferred for linking 21 aerosol with droplet distributions in large scale models.