The Food and Agriculture Organization (FAO), the custodian agency of SDG indicators 6.4.1 and 6.4.2, is running a portal called Water Productivity Open-access portal (WaPOR) to encourage the use of satellite data in crop water use efficiency monitoring. The WaPOR portal currently relies on Terra and Aqua, PROBA-V and Landsat satellite data, many of which are at the end of their expected lifetime. In this project we perform robust, scientific analysis of the possibility to use Copernicus data (especially observations from the Sentinel-2 and Sentinel-3 satellites) in place of the legacy satellites, to produce national maps of evapotranspiration able to capture field-scale crop water use.

The project builds upon methodologies developed during the ESA-funded Sen-ET project. During that project, a robust method was developed to estimate land-surface energy fluxes, including evapotranspiration, at field-scale using Sentinel-2 and Sentinel-3 data, despite the absence of a high-resolution thermal sensor in the Sentinel constellation. The method consisted of a machine-learning data fusion algorithm for sharpening land surface temperature measured by Sentinel-3 at 1 km to the 20 m spatial resolution of Sentinel-2, followed by a physically based modelling of land-surface energy fluxes. In ET4FAO we apply this technique to produce evapotranspiration maps covering the whole of Lebanon and Tunisia at 20 m, 100 m and 300 m spatial resolutions. Those maps are then compared against products available from the FAO WaPOR portal and validated against independent field measurements.