Comparisons of satellite-based models for estimating evapotranspiration fluxes

- Satellite-based models were compared to derive reliable evapotranspiration fluxes.
- The two-source energy balance model performs better in terms fluxes estimation.
- In homogeneous areas the one-source model did not differ from the two-source model.
- The VI-model does not require any up-scaling of the energy fluxes to daily data.

SummaryTwo different types of remote sensing-based techniques were applied to assess the mass and energy exchange process within the continuum soil-plant-atmosphere of a typical Mediterranean crop. The first approach computes a surface energy balance using the radiometric surface temperature (Ts) for estimating the sensible heat flux (H), and obtaining the evapotranspiration fluxes (ET) as a residual of the energy balance. In the paper, the performance of two different surface energy balance approaches (i.e. one-source and two-source (soil + vegetation)) was compared. The second approach uses vegetation indices (VIs), derived from the canopy reflectance, within the FAO-based soil water balance approach to estimate basal crop coefficients to adjust reference ET0 and compute crop ET. Outputs from these models were compared to fluxes of sensible (H) and latent (LE) heat directly measured by the Eddy Covariance method, through a long micrometeorological monitoring campaign carried out in the area of interest.The two-source (2S) model gave the best performance in terms of surface energy fluxes and ET rate estimation, although the overall performance of the three approaches was appreciable. The reflectance-based crop coefficient model has the advantages to do not require any upscaling of the instantaneous ET fluxes from the energy balance models to daily integrated ET. However, its results may be less sensitive to detect crop water stress conditions respect to approaches based on the radiometric surface temperature detection.