The approach of using remote sensing of surface temperature to estimate spatially distributed surface energy balance components is very attractive. This approach has been applied successfully over surfaces with near full vegetation cover. However, large discrepancies between measured and simulated surface fluxes have been observed over surfaces with sparse vegetation cover. The reason for these discrepancies is that the assumption that radiative surface temperature can be equated to aerodynamic surface temperature is not correct over sparsely vegetated surfaces. In this study an empirical model, relating radiative -aerodynamic surface temperature difference to radiative-air temperature gradient and leaf area index, was used to estimate sensible heat flux over sparse shrub in the Central East supersite during the Hydrologic and Atmospheric Pilot Experiment in the Sahel (HAPEX-Sahel) measurement campaign. The result shows that this parameterization leads to reasonable estimates of sensible heat flux; the root mean square error (RMSE) was about 50 W M-2. A second data set over sparse cotton in Arizona had a RMSE of about 20 W m-2. Although the results of this study are encouraging, one should be cautious, however, because there is a need for additional investigation of this procedure.