A Simplified Method for Measuring Land Surface Temperature and Emissivity Using Thermal Infrared Split-window Channels

Satellite land surface temperature (LST) derivation is usually performed at infrared spectrum using multi-channel technique. Because of coupled effects in LST problem, spectral land surface emissivity information is required explicitly or implicitly in such technique, which itself is difficult to be determined accurately. Gillespie et al. [1] presented a numerical method for separating LST and emissivity using the Advanced Spaceborne Thermal Emission and Reflection (ASTER) scanner on NASA’s Earth Observing System (EOS) satellite. Wan et al. [2] developed a two-measurement, multi-band method for deriving the LST and the spectral emissivities simultaneously using spectral radiative transfer equations. Liang [3] further developed an optimization procedure to constraint errors in simultaneous determination process for the LST and the emissivities. All the above methods require intensive numerical computation time since at least 4 to 7 bands data are needed and multiple radiative transfer equations are solved. In this study, we developed a simplified method that separate LST and the spectral emissivities using two split-window algorithms and two measurements at two difference times. The method is tested using the U.S. GOES-8 Imager data and the European Meteosat Second Generation (MSG) Spinning Enhanced Visible and Infrared Imager (SEVIRI) data. Results of the LSTs measured from the GOES-8 Imager is compared with ground LSTs estimated using surface radiation measurements from SURFace RADiation (SURFRAD) budget network.