Cloud optical and microphysical properties, such as optical thickness (Tau) and effective radius (Re), were retrieved by coupled using a nonabsorbing channel (e.g. 0.64 micron of AVHRR), a water-absorbing channel (3.75 micron), and a thermal channel (11 micron). The previous results suggested artificial and natural transformations of cloud properties. Utilization of the 3.75-micron channel was thus a realistic choice for the cloud retrievals from reflected solar radiance. One reason for using the 3.75-micron channel as the water-absorbing channel was that most previous satellite-borne sensors, such as AVHRR, were equipped only with this channel in the near-infrared wavelength domain. However, most modern sensors are equipped with more channels at near-infrared wavelengths. For example, VIRS aboard the TRMM satellite has both 1.6- and 3.7-micron channels; MODIS aboard the Terra satellite and GLI aboard the ADEOS-II satellite have 1.6-, 2.2- and 3.7-micron channels (or nearby). The imaginary indices of refraction of liquid water at these wavelengths are, 10^-5 (1.6-micron), 10^-4 (2.2-micron), and 10^-3 (3.7-micron), and they are still larger than 0.63-micron channelŐs 10^-9, so that 1.6- and 2.2-micron channels, in addition to the 3.7-micron channel are sensitive to cloud droplet sizes. Therefore, it is natural to have interest in retrieving cloud effective radius and investigating the differences among them, if any, by using these extended channels. In this research, we tried to retrieve the global distribution of effective radii by using the 2.1-micron channel of MODIS aboard the Terra as well as 3.7-micron channels and compare them to get the difference of effective radii values. It is noteworthy that using 1.6- and/or 2.2-micron (or nearby) channels has an advantage over using the 3.7-micron channel because the former channels donŐt include undesirable thermal components included in 3.7-micron channels.
The obtained results showed that the ratio Re_2.1 to Re_3.7, has unique features. The ratio reached 2 over low-latitude land areas, but was almost 1 over ocean areas. We considered three reasons for this phenomenon, the effect of sub-pixel size clouds (land contamination), the difference instantaneous field of view (IFOV) of MODIS 2.1- (500m) and 3.7-micron (1 km) channels, and calibration problems. A vertical profile of the cloud droplet size can also influence the results.
Han, Q., W. B. Rossow, and A. A. Lacis, 1994: J. Climate, 7, 465-497.
Kawamoto, K., T. Nakajima, and T. Y. Nakajima, 2001: J. Climate, 14, 2054-2068.
Nakajima T. Y. and T. Nakajima, 1995: J. Atmos. Sci., 52, 4043-4059.