Automatic methods for gas absorption calculation based on correlated k-distribution

• Published in: Journal of quantitative spectroscopy & radiative transfer Vol. 270; p. 107697
• Main Authors: Zhu, Mingwei, Jin, Shuanggen, Tao, Ju, Wu, Xinyue
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2021
• Subjects: Correlated k-distribution model; Finding point method; Gas absorption calculation; Optimization method; Correlated k-distribution model; Finding point method; Gas absorption calculation; Optimization method
• ISSN: 0022-4073; 1879-1352
• DOI: 10.1016/j.jqsrt.2021.107697

•Automatic methods for Gas absorption calculation are developed based on correlated k-distribution•An iterative process and secondary optimization calculation improve the accuracy of results.•The accuracy of two methods is higher than rapid radiative transfer model for general circulation models (RRTMG) Several schemes have been proposed for handling gaseous overlapping bands in the context of the correlated k-distribution model (CKD), but they all require manual operation and the accuracy is limited. In this paper, we proposed two automatic methods for gas absorption calculation based on correlated k-distribution, namely finding point method (FPM) and re-optimized method (ROM), to improve the accuracy and the speed of gaseous absorption calculation. Compared with the line-by-line (LBL) results under standard profiles, the resulting accuracy of FPM is 0.09 Kday−1 in troposphere, -0.3 Kday−1 in stratosphere, -0.18 W/m2 for upward flux and -0.44 W/m2 for downward flux; the accuracy of ROM is 0.1 Kday−1 in troposphere, 0.3 Kday−1 in stratosphere, -0.35 W/m2 for upward flux and -0.18 W/m2 for downward flux. The accuracy of the two methods is higher than rapid radiative transfer model for general circulation models (RRTMG). Under realistic profiles, the accuracy of FPM and ROM is slightly lower than that their accuracy under standard profiles but still higher than RRTMG.