HR: 0800h
AN: A51B-0764    [Abstracts]
TI: A fast radiative transfer model for infrared hyperspectral application to cloudy atmospheres
AU: * Yang, P
EM: pyang@ariel.met.tamu.edu
AF: Texas A&M University, TAMU 3150 Texas A&M University, College Station, TX 77843 United States
AU: NIU, J
AF: Texas A&M University, TAMU 3150 Texas A&M University, College Station, TX 77843 United States
AU: Wei, H
AF: Texas A&M University, TAMU 3150 Texas A&M University, College Station, TX 77843 United States
AU: Huang, H
AF: University of Wisconsin-Madison, West Dayton Street, Madison, WI 53706 United States
AU: Baum, B
AF: University of Wisconsin-Madison, West Dayton Street, Madison, WI 53706 United States
AU: Baum, B
AF: NASA Langley Research Center, NASA Langley Research Center, Hampton, VA 23681 United States
AU: Hu, Y X
AF: NASA Langley Research Center, NASA Langley Research Center, Hampton, VA 23681 United States
AU: Liou, K N
AF: University of California, Department of Atmospheric Sciences, University of California, Los Angeles, CA 90095 United States
AU: Mishchenko, M I
AF: NASA Goddard Institute for Space Studies, NASA Goddard Institute for Space Studies, New York, NY 100025 United States
AU: Strow, L
AF: University of Maryland Baltimore County, fPhysics Department, University of Maryland Baltimore County, Baltimore, MD 21250 United States
AB: When clouds are present, the atmospheric infrared (IR) spectral signatures contain a wealth of information about cloud macrophysical, microphysical, and optical properties. To infer cloud optical thickness and effective particle size from the IR spectral measurements, accurate forward radiative transfer (RT) modeling is required. The conventional approach for the forward RT modeling simulation is based on a combination of the line-by-line model (LBLM) and a rigorous multiple scattering RT method such as the well-known Discrete Ordinates Radiative Transfer (DISORT) model. However, this approach is impractical in terms of CPU requirements for a hyperspectral imager application. Strow et al. (1998) developed a fast clear-sky RT model with very high accuracy when compared to the results from the LBLM. In this study, we develop a fast cloudy-sky RT model to compute the outgoing radiance observed at the top of the atmosphere. The code is capable of dealing with single-layered (plane parallel) clouds as well as a two-layered cloud system, such as an ice cloud overlying a water cloud. For ice clouds, the single-scattering properties of ice crystals are computed from a composite method that is based on the finite-difference time-domain (FDTD) technique, an improved geometric optics method (IGOM), the T-matrix method, and the Lorenz-Mie method. A number of pristine and complex ice crystal habits are considered, and include aggregates, hexagonal columns, hexagonal plates, three-dimensional bullet rosettes, hexagonal hollow columns, spheroids, and droxtals. Based on the ice single-scattering properties, a look-up library is generated of cloud reflectances and transmittances for a range of cloud optical thicknesses, effective particle sizes, and viewing angles. The outgoing radiances can be computed both efficiently and accurately using the pre-computed look-up library. The accuracy of the fast cloudy-sky radiative transfer model has been assessed through comparison with rigorous RT computations based on the LBLM and DISORT models.
DE: 1640 Remote sensing
DE: 0360 Transmission and scattering of radiation
DE: 0649 Optics
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting