AN INVESTIGATION OF CLOUD DISTRIBUTION FROM SATELLITE INFRARED RADIATION DATA

Abstract

A physical-statistical model using Nimbus II MRIR (medium resolution infrared) radiometer channel 2 (10?11 µm) data is formed to estimate cloud cover. In the model, cloud cover is estimated from the probability density distribution of DV2 values defined to be the differences between the earth surface temperatures and the corresponding channel 2 observed equivalent black-body temperatures. For cloud type estimation, channel 1 (6.4?6.9 µm) and channel 2 (10?11 µm) data are used simultaneously in a joint distribution model. Bivariate DV1 and DV2 frequency distributions portray distinct features associated with different cloud patterns from which it was possible to estimate the cloud coverage for different levels in the troposphere. The major limiting factor in estimating cloud coverage in this study was the areal resolution of the MRIR radiometer. The results indicate that a reliable global climatology may be formed, using higher resolution radiometers combined with irradiance information from other spectral intervals.