Application of a Remote Sensing Method for Estimating Monthly Blue Water Evapotranspiration in Irrigated Agriculture

In this paper we show the potential of combining actual evapotranspiration (ET_actual) series obtained from remote sensing and land surface modelling, to monitor community practice in irrigation at a monthly scale. This study estimates blue water evapotranspiration (ET_b) in irrigated agriculture in two study areas: the Horn of Africa (2010–2012) and the province of Sichuan (China) (2001–2010). Both areas were affected by a drought event during the period of analysis, but are different in terms of water control and storage infrastructure. The monthly ET_b results were separated by water source—surface water, groundwater or conjunctive use—based on the Global Irrigated Area Map and were analyzed per country/province. The preliminary results show that the temporal signature of the total ET_b allows seasonal patterns to be distinguished within a year and inter-annual ET_b dynamics. In Ethiopia, ET_b decreased during the dry year, which suggests that less irrigation water was applied. Moreover, an increase of groundwater use was observed at the expense of surface water use. In Sichuan province, ET_b in the dry year was of similar magnitude to the previous years or increased, especially in the month of August, which points to a higher amount of irrigation water used. This could be explained by the existence of infrastructure for water storage and water availability, in particular surface water. The application presented in this paper is innovative and has the potential to assess the existence of irrigation, the source of irrigation water, the duration and variability in time, at pixel and country scales, and is especially useful to monitor irrigation practice during periods of drought.