Anthony D Del Genio
NASA/Goddard Institute for Space Studies
Code 611
2880 Broadway
New York, NY 10025

E-mail: anthony.d.delgenio@nasa.gov
Phone: 212-678-5588
Fax: 212-678-5622

Climate GCM predictions of cloud feedback depend on their parameterizations of the rate at which condensate is removed from clouds as precipitation and the effect of this depletion of large particles on the radiative properties of the cloud material that remains. The combination of CloudSat and CALIPSO data with existing information from the Aqua satellite offers an unprecedented opportunity to evaluate the processes in global climate models that determine the effect of major cloud systems on the global energy and hydrologic cycle. We propose to focus on two types of pervasive vertically developed cloud phenomena in the climate system:

1. Tropical cumulus anvils and associated cirrus. We will use nearly colocated CloudSat CPR ice water content cross-sections and Aqua AMSR-E precipitation rates to distinguish the precipitating and non-precipitating segments of cumulus anvils, the characteristic radiative heating-cooling rate profiles through each, and the implied mesoscale vertical velocity profiles within the anvils. If a CloudSat simulator is made available to Science Team members, we will also use the diagnosed drop size distribution for cumulus updrafts and anvils currently implemented in the GISS GCM to simulate radar reflectivity profiles and perform a direct comparison with CloudSat reflectivity profiles. We will also track simultaneous Aqua CERES albedos along the CloudSat track. The combination of IWC, radar reflectivity, albedo, and rainfall information will be used to improve the physics underlying the GCM's predictions of updraft speeds, ice crystal fallspeeds, and drop size distribution. Finally, the early afternoon A-train snapshot of these systems will be placed into the larger context of the complete diurnal cycle using simultaneous ISCCP data on system size and diurnal cloud property variation, and GERB diurnally-resolved TOA radiation budget information over the Atlantic, Europe and Africa.
2. Midlatitude baroclinic storms. Synoptic storms in midlatitude winter are under-resolved by climate GCMs, leading them to overpredict high optically thick clouds and underpredict cirrus in such systems. We plan to use ERA-40 reanalysis products, analyzed with an existing baroclinic storm identification algorithm, to identify synoptic storm occurrence along the A-train track. CloudSat cloud boundary information will be aggregated to produce the first truly three-dimensional composite baroclinic storm cloud structure. The tilts of frontal cloud systems associated with their ageostrophic circulations and ambient wind shear will be determined and compared to climate GCM composite structures of the same events. Futhermore, since cloud phase often changes from base to top in synoptic storms, we plan to use CALIPSO cloud top altitude and depolarization data to systematically map the relationship of cloud top phase to cloud top temperature and assess the validity of analogous retrievals made by Aqua MODIS. We will especially look to see whether differences in cloud vertical structure between Pacific and Atlantic storm track events explain differences in cloud phase behavior that appear in MODIS data. AMSR-E precipitation retrievals will be composited over the storm structure as well and be used to determined threshold ice or liquid water contents for precipitation.

The GISS GCM, whose standard configuration has 32 layers, also exists in an experimental version with 100 layers, giving it tropospheric vertical resolution of 100-300 m, comparable to the CloudSat vertical bin spacing. By using both models we will be able for the first time to directly evaluate the effect of representing cloud internal structure on our ability to simulate cloud evolution and heating/cooling rates.