DETERMINATION OF SENSIBLE HEAT-FLUX OVER SAHELIAN FALLOW SAVANNA USING INFRARED THERMOMETRY

The estimation of the partitioning of available energy from remote sensing techniques is addressed for a Sahelian fallow savannah. It is a composite vegetation consisting of shrubs of Guiera senegalensis scattered above a stand of sparse grass. A two-layer model is employed to estimate sensible heat flux (H) from radiometric surface temperature, with data collected during the HAPEX-Sahel international experiment, carried out in Niger in 1992. The model, based upon the assumption that the radiometric surface temperature (T(r)) might be represented by the composite surface temperature (area-weighted mean of shrub and grass temperatures), leads to a simple formulation of H as a function of the temperature difference between the surface and the air (T(r) - T(a)) and the temperature difference between the grass and the shrubs deltaT. The estimates of the model compare fairly accurately with measurements obtained by the Bowen ratio-energy balance method, the root mean square error being about 52 W m-2. Because deltaT is not easily measured from remote sensing systems, it has been shown that for the fallow savannah this temperature difference is linearly correlated to (T(r) - T(a)) with r2 = 0.94. Therefore, it is possible to estimate sensible heat flux from (T(r) - T(a)) without additional component temperature measurements.