An Accurate And Efficient Radiative Transfer Model For Satellite Radiometer Images And Its Application In Model Evaluation

With the development of satellite technologies and various satellite observations,the rigorous radiative transfer models(RTM)are impractical to use for satellite related applications because of the huge amount of independent calculations.In this study,we develop an accurate and efficient RTM for simulating all-sky images from Fengyun satellite radiometers,namely FYRTM(RTM for Fengyun radiomter).Moreover,based on the advantage of specific designed RTM,a "Radiance-based" evaluation approach is introduced,and the capability of weather numerical predict model,as well as the cloud and atmospheric characteristics in different atmospheric reanalyses are studied.In this study,the pre-computed cloud and aerosol multiple scattering and emission look-up tables help to improve the computational efficiency of radiative transfer.The correlated k-distribution models are developed to for atmospheric transmittance and the surface albedo are accelerated with pre-computed look-up tables.The developed FYRTM has been evaluated against a rigorous simulation based on discrete ordinate radiative transfer model(DISORT),and the results indicate that FYRTM is two to three orders of magnitudes faster than the DISORT-based simulations.Compared to the rigorous model,FYRTM relative errors are within 2%at solar channels,and brightness temperature(BT)differences are with 1 K at infrared channels.Comparison with the popular fast Community Radiative Transfer Model(CRTM)indicates that FYRTM is computationally similar at solar channels,but three times faster at infrared channels.Furthermore,simulated reflectances/BTs using FYRTM are in a good agreement with the satellite observations.It is expected to be used in the study of calibration,satellite retrieval,assimilations and so on.To avoid the influence of uncertainties from satellite retrieval products,a radiance-based evaluation is introduced and performed to evaluate and analyze the capability of Weather Research and Forecasting model(WRF).In this approach,the fast forward radiative transfer simulations are used as a bridge to translate the model simulated parameters to corresponding radiances that are expected to be observed by satellites.By direct comparing the simulated infrared BTs with observations from the Visible and Infrared Spin-Scan Radiometer onboard Fengyun-2E satellite,the WRF simulation has solid capability to capture the atmosphere and cloud characteristics broadly and reproduces the high-level clouds well,whereas the low-to-midlevel clouds are underestimated in the case study.We also analyze the quality of cloud and atmospheric properties from reanalysis datasets:China Meteorological Administration reanalysis(CRA),ECMWF's Fifth-generation reanalysis(ERA5),and Modern-Era Retrospective analysis for Applications,Version 2(MERRA-2).Observations from Advanced Himawari Imager onboard the Himawari-8 satellite are used as the "truth".Generally,CRA,ERA5,and MERRA-2 are all capable of representing the atmospheric and cloud characteristics over the region.Larger statistical errors occur over the oceanic region around the Equator and areas with complex surface features.Analysis indicates that ERA5 achieves the generally best accuracy over the entire observed region.Results of CRA also reveal reasonable simulations,and they are close to those in ERA5,whereas for MERRA-2,the deviations are slightly larger because of largely overestimated clouds.Overall,with demonstrated accuracy and efficiency,it is believed that the particularly designed RTM,i.e.,the FYRTM,is expected to play an important role in developing algorithms to retrieve various atmospheric products from Fengyun measurement and some other fields.Meanwhile,we believe that the radiance-based evaluation approach discussed in the study also be able to help us to understand the model simulated cloud and atmospheric properties.