Implementation of an ice crystal single-scattering property database in the radiative transfer model SCIATRAN

• Published in: Journal of quantitative spectroscopy & radiative transfer Vol. 253; p. 107118
• Main Authors: Pohl, Christine, Rozanov, Vladimir V., Mei, Linlu, Burrows, John P., Heygster, Georg, Spreen, Gunnar
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2020
• Subjects: Generalized spherical functions; Habit mixtures; Ice crystal single-scattering properties; Particle size distribution; Radiative transfer; Habit mixtures; Ice crystal single-scattering properties; Particle size distribution; Radiative transfer; Generalized spherical functions
• ISSN: 0022-4073; 1879-1352
• DOI: 10.1016/j.jqsrt.2020.107118

•Single-scattering properties of nine ice crystals are implemented in the radiative transfer (RT) model SCIATRAN.•The scattering matrix is expanded in generalized spherical functions.•A gamma size distribution and mixing ratios consider the polydispersity and habit mixtures in ice particulate media.•The implemented database enables the RT calculation through ice particulate media. Ice crystal single-scattering properties (single-scattering albedo, extinction cross section, scattering matrix) determine the radiative transfer through ice clouds and snow layers. In the radiative transfer model SCIATRAN, only a limited range of particle sizes and shapes were available. Consequentially, and to achieve a much more realistic representation, a new database of nine ice crystals (droxtals, solid and hollow columns, aggregates of eight hexagonal solid columns, hexagonal plates, aggregates of five or 10 hexagonal plates, and solid and hollow bullet rosettes) with maximum dimensions from 2 to 10 000 µm has been implemented and tested. SCIATRAN now includes the relevant single-scattering albedo, extinction cross section, and scattering matrix. Additionally, the scattering matrix is expanded in generalized spherical functions based on a combination of trapezoidal and Gaussian quadrature. This implemented database in conjunction with a built in user-defined particle size distribution and an arbitrary habit ratio enable the radiative transfer through a variety of layers, composed of polydisperse ice particles having particular habit mixtures. A comparison of satellite measured snow surface reflectances with respective simulations demonstrates that assuming ice crystal habit mixtures rather than single-habits as snow crystals reduces significantly the differences between the model and satellite data from 8.5 % to 2 %.