Estimating rice water requirements in the Lis Valley (Portugal) using remote sensing platforms: preliminary results for the 2020 cultivation season

In irrigated agricultural areas, where the availability of water for irrigation does not rely on any water storage, water management requires special attention, in particular under large annual and inter-annual variability in the hydrological regime and the uncertainty of climate change. The inherent increased vulnerability of the agro-ecosystem, makes the monitoring of crop conditions and water requirements a valuable tool for improving water use efficiency and, therefore, crop yields.

This presentation focus on one such agricultural area, located in the Lis Valley (Centre of Portugal), which is a rather vulnerable area also facing drainage and salinity problems. The study aims at contributing to better characterizing the temporal and spatial distribution of rice water requirements during the growing season. Irrigation water sources are the Lis River and its tributaries, which discharges depend directly from precipitation. The most important problems of water distribution in the Lis Valley irrigation district are water shortage and poor water quality in the dry summer period, aggravated by limitations of the irrigation and drainage systems that date back to the end of the 1950’s.

We report preliminary results on using remote sensing data to better understand rice cropping local conditions, obtained within project GO Lis (PDR2020-101-030913) and project MEDWATERICE (PRIMA/0006/2018). Rice irrigation is traditionally conducted applying continuous flooding, which requires much more irrigation water than non-ponded crops, and therefore needs special attention. In particular, data obtained from satellite Sentinel-2A land surface imagery are compared with data obtained using an unmanned aerial vehicle (UAV). Data for rice cultivated areas during the 2020 cultivation season, together with weather and crop parameters, are used to calculate biophysical indicators and indices of water stress in the vegetation. Actual crop evapotranspiration was appraised with remote sensing based estimates of the crop coefficient (Kc) and used to assess rice water requirements. Procedures and methodologies to estimate Kc were tested, namely those based on vegetation indices such as the Normalized Difference Vegetation Index (NDVI). Results are discussed bearing in mind the usefulness of the diverse tools, based on different resolution data (Sentinel-2A and UAV), for improving the understanding of the impacts of irrigation practices on crop yield and main challenges of rice production and water management in the Lis Valley irrigation district.