A New Method for Fast Computation of Three-Dimensional Atmospheric Infrared Radiative Transfer in a Nonscattering Medium, with an Application to Dynamical Simulation of Radiation Fog in a Built Environment

Abstract

Abstract The atmospheric radiation field has seen the development of more accurate and faster methods to take into account absorption. Modeling fog formation, where infrared radiation is involved, requires accurate methods to compute cooling rates. Radiative fog appears under clear-sky conditions owing to a significant cooling during the night where absorption and emission are the dominant processes. Thanks to high-performance computing, high-resolution multispectral approaches to solving the radiative transfer equation are often used. Nevertheless, the coupling of three-dimensional radiative transfer with fluid dynamics is very computationally expensive. Radiation increases the computation time by around 50% over the pure computational fluid dynamics simulation. To reduce the time spent in radiation calculations, a new method using analytical absorption functions fitted by Sasamori on Yamamoto’s radiation chart has been developed to compute an equivalent absorption coefficient (spectrally integrated). Only one solution of the radiative transfer equation is needed against Nband × Ngauss for an Nband model with Ngauss quadrature points on each band. A comparison with simulation data has been done and the new parameterization of radiative properties proposed in this article shows the ability to handle variations of gas concentrations and liquid water.