Premium
Joint analysis of cloud top heights from CloudSat and CALIPSO: New insights into cloud top microphysics
Author(s) -
Hagihara Yuichiro,
Okamoto Hajime,
Luo Zhengzhao Johnny
Publication year - 2014
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1002/2013jd020919
Subject(s) - lidar , environmental science , pathfinder , cloud top , cloud computing , cloud fraction , satellite , atmospheric sciences , meteorology , remote sensing , cloud cover , geology , computer science , geography , physics , astronomy , library science , operating system
We examined the differences in the cloud top heights (CTHs) detected by the CloudSat radar and Cloud‐Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) lidar. Theoretical estimates have shown that CloudSat has higher sensitivity than CALIPSO does when large particles exist. In such case it might be possible that CloudSat‐determined CTHs are larger than CALIPSO‐determined CTHs. We compared the global distribution of CTHs detected by CloudSat and CALIPSO (version 3, V3) using our cloud mask schemes after carefully selecting data during September–November 2006. The global mean fraction of clouds where CloudSat‐determined CTHs were larger than CALIPSO‐determined CTHs turned out to be unexpectedly large. The fractions were 26% and 39% at low level and midlevel, and the corresponding CTH differences were 0.56 km and 0.86 km, respectively. On the western coasts of continents, these clouds occurred within temperature inversions. Accounting for the differences in sensitivity to particle size between CloudSat and CALIPSO, the existence of such clouds indicates that the cloud tops consist of large particles with small number concentration. The discovery of such clouds was revealed by our joint analysis of CloudSat and CALIPSO. When the standard vertical feature mask (VFM) V3 was used, these clouds were also found but the fractions were less pronounced. The differences were partly attributed to the overestimation of cloud fraction in the VFM V3, although the degree of misidentification in V3 was reduced compared with that of V2.