Droplet Growth in Warm Water Clouds Observed by the A-Train. Part II: A Multisensor View

Abstract Hydrometeor droplet growth processes are inferred from a combination of Aqua/Moderate Resolution Imaging Spectroradiometer (MODIS) cloud particle size observations and CloudSat/Cloud Profiling Radar (CPR) observations of warm water clouds. This study supports the inferences of a related paper (Part I) (i) that MODIS-retrieved cloud droplet radii (CDR) from the 3.7-μm channel (R37) are influenced by the existence of small droplets at cloud top and (ii) that the CDR obtained from 1.6- (R16) and 2.1-μm (R21) channels contain information about drizzle droplets deeper into the cloud as well as cloud droplets. This interpretation is shown to be consistent with radar reflectivities when matched to CDR that were retrieved from MODIS data. This study demonstrates that the droplet growth process from cloud to rain via drizzle proceeds monotonically with the evolution of R16 or R21 from small cloud drops (on the order of 10–12 μm) to drizzle (CDR greater than 14 μm) to rain (CDR greater than 20 μm). Thus, R16 or R21 is an indicator of hydrometeor droplet growth processes whereas R37 does not contain information about coalescence. A new composite analysis, the contoured frequency diagram, is introduced to combine CloudSat/CPR reflectivity profiles and reveals a distinct trimodal population of reflectivities corresponding to cloud, drizzle, and rain modes.