Benefits and pitfalls of irrigation timing and water amounts derived from satellite soil moisture

Summary

This study evaluated two satellite-based approaches for retrieving irrigation timing and water amounts from Sentinel-1 soil moisture data across the 83,000 km² Ebro basin in Spain. Both the SM_Delta method (comparing pixel soil moisture to surrounding rainfed areas) and the SM_Inversion method (estimating total water input minus precipitation) showed reasonable agreement with ground-truth irrigation data at district scale, though both erroneously detected irrigation over some rainfed pixels. The findings suggest these satellite approaches are viable for monitoring irrigation at regional scales when auxiliary land-use information is available, though further refinement is needed for field-level discrimination.

Regional applicability

The study was conducted in Spain and results may transfer to United Kingdom irrigation contexts with similar soil types and climatic zones, though UK irrigation is less intensive than in the semi-arid Ebro basin. The methods' reliance on Sentinel-1 data offers global applicability, but accuracy depends on local soil moisture variability and availability of ancillary irrigated-area maps, which may be better established in UK agri-environmental monitoring schemes.

Key measures

Irrigation timing agreement, irrigation water volume estimates (m³), Pearson correlation coefficient (R), bias, spatial and temporal irrigation patterns, discrimination accuracy for irrigated vs. rainfed pixels

Outcomes reported

The study assessed two satellite-based algorithms (SM_Delta and SM_Inversion) for estimating irrigation timing and water volumes using Sentinel-1 soil moisture data across the Ebro basin in Spain during 2017–2019. Both methods showed satisfactory agreement with district-scale reference irrigation data (Pearson R of 0.67 and 0.71) but were unsuitable for discriminating irrigated from rainfed fields without auxiliary information.

Topic tags