An estimation method for regional sensible heat flux on vegetation using satellite infrared temperature

Spacial distribution of sensible heat flux on rice paddy of which leaf area index (LAI) range is 0.01 through 5, is evaluated in regional scale together with possible error estimation. It is found that three types of correction of satellite infrared temperature (correction for atmospheric absorption, satellite viewing angle, and emissivity) are required to evaluate sensible heat flux on vegetation surface as rice paddy. Root mean square error of the post corrected satellite temperature is about 1degC and that of resulted sensible heat flux is about 25 Wm^-2 compared with ground truth. The amount of correction is 2 through 4degC for atmospheric absorption and emissivity, and -5 through 0degC, depended on LAI and insolation, for satellite viewing angle. The correction of satellite viewing angle is critical for sensible heat evaluation under LAI range being 0.2 through 2 and daytime conditions. If the correction of satellite viewing angle is not performed, RMS error would be doubled.

A new parameterization is proposed for the viewing angle correction based on the ground truth data. This parameterization is nonlinear function of difference between air and surface infrared temperature, and the amount of viewing angle correction is not approximately depended on the viewing angle itself. Difference of errors of estimated sensible heat using between the viewing angle correction with aerodynamic bulk transfer coefficient and the radiometric roughness parameter is not significant when the magnitude of sensible heat flux is not quite large. Using the radiometric roughness parameter, systematic bias could cause in case of sensible heat flux being over about 150 Wm^-2 because of the nonlinearity of the amount of the viewing angle correction, while the linear relation is assumed in the radiometric roughness concept.