paNTICA: A Fast 3D Radiative Transfer Scheme to Calculate Surface Solar Irradiance for NWP and LES Models

Abstract

he resolution of numerical weather prediction models is constantly increasing, making it necessary to consider three-dimensional radiative transfer effects such as cloud shadows cast into neighboring grid cells and thus affecting radiative heating. For that purpose, fast approximations are needed since three-dimensional radiative transfer solvers are computationally far too expensive. For the solar spectral range, different approaches of how to consider three-dimensional effects were presented in the past?in particular, the tilted independent column approximation (TICA), which aims at improving the calculation of the direct radiation, and the nonlocal tilted independent column approximation (NTICA), which is used to additionally correct the diffuse radiation. Here a new version of NTICA is presented that?in contrast to earlier approaches?is applicable for a variety of cloud scenes and grid resolutions and for arbitrary solar zenith angles. This new parameterization for the diffuse irradiance is then applied to the two different TICA approaches and the results are compared with a full 3D Monte Carlo calculation. It is shown that both approaches strongly improve the calculation of radiation fluxes if the new parameterization for the diffuse irradiance?what the authors call ?parameterized NTICA (paNTICA)??is applied. It is found that the method in which TICA is only applied to direct radiation yields the better results. The studies show that consideration of three-dimensional effects is inevitable if higher model resolutions are used in the future. This paper proposes ways to consider these effects and, thus, to substantially reduce the errors made with one-dimensional radiative transfer solvers.