Infrared Radiative Transfer Through a Regular Array of Cuboidal Clouds

View More View Less
  • 1 Department of Meteorology, University of Maryland, College Park 20742and Laboratory for Atmospheric Sciences, Goddard Space Flight Center, Greenbelt, MD 20771
  • | 2 Space Science and Engineering Center, University of Wisconsin, Madison 53706
© Get Permissions
Full access

Abstract

A study has been made of infrared radiative transfer through a regular array of cuboidal clouds which considers the interaction of the sides of the clouds with each other and the ground. The theory is developed for black clouds and is extended to scattering clouds using a variable azimuth two-stream (VATS) approximation (Harshvardhan et al., 1981). It is shown that geometrical considerations often dominate over the microphysical aspects of radiative transfer through the clouds. For example, the difference in simulated 10 μm brightness temperature between black isothermal cubic clouds and cubic clouds of optical depth 10, is <2 K for zenith angles <50° for all cloud fractions when viewed parallel to the array.

The results show that serious errors are made in flux and cooling rate computations if broken clouds are modeled as planiform. Radiances computed by the usual practice of area-weighting cloudy- and clear-sky radiances are in error by 2–8 K in brightness temperature for cubic clouds over a wide range of cloud fractions and zenith angles. It is also shown that the lapse rate does not markedly affect the exiting radiances for cuboidal clouds of unit aspect ratio and optical depth 10.

Abstract

A study has been made of infrared radiative transfer through a regular array of cuboidal clouds which considers the interaction of the sides of the clouds with each other and the ground. The theory is developed for black clouds and is extended to scattering clouds using a variable azimuth two-stream (VATS) approximation (Harshvardhan et al., 1981). It is shown that geometrical considerations often dominate over the microphysical aspects of radiative transfer through the clouds. For example, the difference in simulated 10 μm brightness temperature between black isothermal cubic clouds and cubic clouds of optical depth 10, is <2 K for zenith angles <50° for all cloud fractions when viewed parallel to the array.

The results show that serious errors are made in flux and cooling rate computations if broken clouds are modeled as planiform. Radiances computed by the usual practice of area-weighting cloudy- and clear-sky radiances are in error by 2–8 K in brightness temperature for cubic clouds over a wide range of cloud fractions and zenith angles. It is also shown that the lapse rate does not markedly affect the exiting radiances for cuboidal clouds of unit aspect ratio and optical depth 10.

Save