It is of great interest to investigate the optical, microphysical, and geometrical properties of clouds that play
crucial role in the earth climate system. Water clouds are generally optically thick and consequently have a
great cooling effect on earth-atmosphere radiation budget. The water clouds usually exist in a lower troposphere
where aerosol-cloud interaction occurs frequently, and then cloud droplet size variation in
solar radiation as well. Further, a cloud layer height is one of the key properties that determine downward
longwave radiation and then surface radiation budget. In this study, top height, geometrical thickness and
bottom height of a water cloud layer are investigated as cloud geometrical properties in particular.
Several studies show that observation data of some spectral regions including oxygen A-band, enables us to
retrieve the cloud geometrical properties as well as the optical thickness and the effective particle radius. In
this study, an algorithm was developed to retrieve simultaneously the cloud optical thickness, effective particle
radius, top height, geometrical thickness and then bottom height of a cloud layer with the spectral observation
of visible, near infrared, thermal infrared, and oxygen A-band channels.
This algorithm was applied to Advanced Earth Observing Satellite － II / Global Imager (ADEOS-II / GLI)
dataset so as to retrieve global distribution of cloud geometrical properties. The retrieved results around Japan
were compared with other observation such as ground-based active sensors, which suggest this algorithm works
for cloud system over ocean at least although the cloud bottom height was underestimated. The underestimation
is attributed to cloud inhomogeneity at this stage and should be investigated in detail.