Remote Sensing of the Thickness and Composition of Cirrus Clouds from Satellites

Kuo-Nan Liou Department of Meteorology, University of Utah, Salt Lake City 84112

Search for other papers by Kuo-Nan Liou in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A retrieval technique is presented for the determination of the surface temperature, the thickness and transmissivity of cirrus clouds, and the fraction of the cirrus cloudiness by means of four observed upwelling radiances in the 10 μm window region. On the basis of radiative transfer calculations for mean wavenumbers of 900, 950, 1100 and 1150 cm−1, assumptions are made in the theoretical retrieval analyses that water vapor effects above cirrus clouds are negligible and that ratios of the transmissivities are linear functions of the cloud thickness. Error analyses employing climatological data reveal that independent random errors in temperature and humidity profiles introduce insignificant errors in the four resulting parameters. The resulting errors caused by random errors in the expected upwelling radiances, however, depend upon their standard deviations. Once the thickness and the transmissivity at a given wavenumber of a cirrus cloud have been determined, we illustrate that the vertical ice content may be estimated assuming that ice particles are randomly oriented in a horizontal plane.

Abstract

A retrieval technique is presented for the determination of the surface temperature, the thickness and transmissivity of cirrus clouds, and the fraction of the cirrus cloudiness by means of four observed upwelling radiances in the 10 μm window region. On the basis of radiative transfer calculations for mean wavenumbers of 900, 950, 1100 and 1150 cm−1, assumptions are made in the theoretical retrieval analyses that water vapor effects above cirrus clouds are negligible and that ratios of the transmissivities are linear functions of the cloud thickness. Error analyses employing climatological data reveal that independent random errors in temperature and humidity profiles introduce insignificant errors in the four resulting parameters. The resulting errors caused by random errors in the expected upwelling radiances, however, depend upon their standard deviations. Once the thickness and the transmissivity at a given wavenumber of a cirrus cloud have been determined, we illustrate that the vertical ice content may be estimated assuming that ice particles are randomly oriented in a horizontal plane.

Save