In arid and semi-arid regions, efficient water resource management depends on the accurate estimation of actual evapotranspiration (ET), a critical parameter for determining crop water needs. However, this estimation is challenging due to complex soil–vegetation–atmosphere interactions and the scarcity of reliable in situ data.
This study evaluates the use of the FAO-56 dual crop coefficient approach, which separates ET into basal crop (Kcb), soil evaporation (Ke), and water stress (Ks) coefficients. We investigate the feasibility of estimating these coefficients using freely available satellite-derived variables: Normalized Difference Vegetation Index (NDVI), land surface temperature (LST), and surface soil moisture at 5 cm depth (SSM). This remote sensing-based method addresses data limitations while ensuring reliable ET estimation. To assess its practical applicability, the methodology was tested on two wheat fields in the Haouz plain of Morocco during the 2016/2017 and 2017/2018 growing seasons, under contrasting irrigation regimes—one fully irrigated and the other experiencing water stress.
The results show that the FAO-56 coefficients can be accurately estimated from satellite data. The Kcb coefficient correlated strongly with Sentinel-2-derived NDVI (R² = 0.70), while Ke showed high correlation with SSM from Sentinel-1 (R² = 0.81). The Ks coefficient was derived from a thermal index based on Landsat LST, using reference values under stressed and well-watered conditions. ET estimates derived from these parameters were validated against eddy covariance measurements, showing strong agreement: R² = 0.77 (0.87) and RMSE = 0.68 mm (0.69 mm) for 2016/2017, and R² = 0.74 (0.70) and RMSE = 0.37 mm (0.45 mm) for 2017/2018, for the stressed and non-stressed plots, respectively.
These findings demonstrate the potential of satellite data for reliably estimating FAO-56 parameters, offering a scalable and cost-effective solution for ET monitoring and precision irrigation in data-limited, climate-vulnerable regions.
