Retrieval of Infrared Land Surface Emissivity and their Impacts on Satellite Data Assimilation

The emissivity is one of the most important basis data in retrieval of land surface temperature (LST) with generalized split window method. And more accurate surface emissivity is required in the radiative transfer models for the assimilation of satellite based radiation data into the NWP models. Infrared land surface emissivity contains high spatio-temporal variability caused by the surface material, wavelength, temperature, soil-moisture, and so on. However, almost of the surface emissivity retrieved in previous studies focused on spectral resolution (high/ultra spectral) and assume the emissivity does not change within a month. In this study, the land surface emissivity on infrared window channels (IR 11µ and 12µ) over the Asian region are retrieved using simple ratio method between ground and vegetation. The 8-day frequency NDVI based surface emissivity dataset shows the temporal/seasonal variation according to the growth and scenescence of the vegetation. Currently, infrared emissivity over land surface in fast radiative transfer model is prescribed as constant. It will be replaced with spatially and temporally variated surface emissivity according to the land cover types and vegetation status. And the impact of improved land surface emissivity to the data assimilation of satellite radiation will be investigated.