Bennartz, R (2007). Global assessment of marine boundary layer cloud droplet number concentration from satellite. JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 112(D2), D02201.

Abstract
Global satellite data are used to infer the droplet number concentration of marine boundary layer clouds. In a first step, two and a half years (July 2002 to December 2004) of Moderate Resolution Imaging Spectroradiometer (MODIS, on board NASA's Aqua satellite) level 3 data estimates of cloud effective radius and optical thickness are used to derive cloud droplet number concentration N and cloud geometrical thickness H under the assumption of adiabatically stratified clouds. Theoretical error estimates show that for a liquid water path higher than 30 g/m(2) and a cloud fraction higher than 0.8, H (N) can be derived with a relative uncertainty of better than 20% (80%). To further validate the estimates of N and H, cloud liquid water path is calculated and compared to independent observations of cloud liquid water path from the passive microwave Advanced Microwave Scanning Radiometer (AMSR-E), also on board Aqua. Good agreement between the two different data sets is found. In a second step, the global distribution of cloud droplet number concentration in stratiform boundary layer clouds is evaluated. The data are separated into observations that are likely to be drizzling and drizzle- free using published relations between drizzle rate and N and H. It is found that the mean droplet number concentration over remote Northern Hemisphere oceans is higher than over the southern oceans (64-89 cm(-3) in the Northern Hemisphere and 40-67 cm(-3) in the Southern Hemisphere). Leeward of the major continents cloud droplet number concentration is generally high with maximum values close to the coasts. On the Southern Hemisphere, especially over the east Pacific Ocean, microphysical conditions were almost constant through the entire observation period. Over the southeast Atlantic Ocean, the cloud microphysical variability appears to be strongly influenced by the dry biomass burning season in Africa.

DOI:
10.1029/2006JD007547

ISSN:
0148-0227