Impact of temperature dependency of ice scattering properties on spectrally resolved radiance at far infrared wavelengths

The ability to estimate the longwave radiative effects of ice clouds relies on the accuracy in modelling their physical and radiative properties at mid and far infrared (MIR and FIR) wavelengths. In the FIR part of the spectrum (15-100 µm), the optical constants of water ice depend on temperature. This spectral dependency affects the scattering properties of ice clouds and consequently, the radiation measured both at the surface and at the top of the atmosphere. Recently, advances in modelling of cloud radiative properties resulted also in the assembly of single scattering single particle dataset with temperature dependency. Ren et al.1 set-up a new dataset of optical properties for severely surface-roughened 8-piece hexagonal column aggregate ice particles in the far infrared and for temperature ranging from 160 K to 270 K. The assumed particle shape is the same included in a popular database of ice crystal radiative properties (Yang et al.2) computed for a single temperature (266 K) and adopted in many retrieval algorithms such as the one for the Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6. In this work, we perform multiple scattering line by line computations to evaluate the impact of the temperature on high-resolution upwelling (at the top of the atmosphere) and downwelling (at the surface) radiances in the presence of ice clouds. To assess the level of accuracy of the simulated radiance fields, in the assumption that no temperature dependency is needed, simulations are run in the far infrared part of the spectrum for multiple scenarios provided by the Numerical Weather Prediction Satellite Facility (NWP SAF) model profile d ata t hat a re r epresentative o f t he t ypical v ariability a nd e xtremes of atmospheric conditions. The impact of the new ice temperature dependent dataset on outgoing and surface radiation at FIR, are analysed for multiple ice cloud cases and related to the assumed cloud and atmospheric properties of the observational conditions. It is found that in about one third of the analysed cases the differences are remarkable in comparison to the noise level of the future Far-infrared Outgoing Radiation Understanding and Monitoring (FORUM) satellite mission which will be able to spectrally resolve the FIR band. The ice cloud altitude (i.e. temperature) and the water vapor content between the observer and the cloud, play a major role in the significance of the difference between using a temperature dependent or independent cloud properties dataset.