1. Introduction
Surface temperature is a key variable for monitoring land surface energy balance and in particular evapotranspiration ([2], [4], [5], [21], [28], [26], [24]). Remote measurement of surface temperature allows assessing surface energy balance at various spatial scales from satellite and airplane platforms or from hand-held thermal infrared radiometers. However, surface temperature cannot be directly derived from thermal measurements. Measured radiation includes not only the radiation emitted by the surface but also the radiation emitted by the atmosphere. If we consider top of canopy measurements, this additional radiation results from the reflection of atmospheric radiation by the surface toward the sensor. Impact of surface emissivity must also be accounted since it directly affects the level of emitted radiation at a given temperature. Poor knowledge in either surface emissivity or atmospheric radiation results in error in the determination of surface temperature from remote sensing measurement. These effects have been recognized for a long time (e.g. [5], [15], [25]). Olioso (1995b, [20]) showed that an error of on surface emissivity results in an error between 0.6 and 0.9 K on surface temperature. [20] also reported that this error strongly depends on the way atmospheric radiation is characterized and that it is very important to consider the atmospheric radiation in the same spectral range as the sensor (e.g. . This is particularly problematic since measuring the atmospheric radiation in a limited spectral range is complex.