Abstract
Recent infrared sounders such as the Atmospheric Infrared Sounder and the Infrared Atmospheric Sounding Interferometer (IASI) have thousands of channels. Since the channels have fewer degrees of freedom than the number of channels, the channels are not independent, and thus correlations are expected among them. On examination of the transmittances of IASI channels, we found that many combinations of channels have high statistical correlations. Furthermore, the transmittances of 387 of the 616 IASI channels are statistically predictable from only 41 channels. An empirical estimation of transmittances by using a fast radiative transfer model greatly improved the computational efficiency of an atmospheric sounder.
