An integrated satellite remote sensing approach for mapping field-scale irrigation dynamics to support water management in highly regulated semi-arid catchments

Jackline W. Muturi, Sayantan Majumdar, Christopher E. Ndehedehe, Nathan O. Agutu, Mark J. Kennard

Accurate monitoring of the spatio-temporal extent of irrigated croplands is critical for effective water management. Because vegetation greenness alone cannot distinguish irrigation from rainfall-driven growth, satellite-based detection of irrigated systems remains challenging. This study applies a simple, yet robust thresholding approach based on the fraction of reference evapotranspiration (ET 0 F) and enhanced vegetation index (EVI) to discriminate irrigated areas from rainfed croplands in the Namoi catchment, from 2019 to 2025. The objectives were to quantify the spatio-temporal variability of irrigated cropland, compare results with existing historical irrigated area datasets, and validate the performance of the ET 0 F-EVI thresholding approach. Irrigated extents varied substantially among years, with irrigation predominantly clustered in the central and south-eastern parts of the catchment, and only limited areas exhibiting persistent irrigation. Validation across 40 fields showed that the approach behaves as a conservative classifier with a moderate accuracy (61.4%) and specificity (79.9%), but with low sensitivity (38%) and low classification agreement (Kappa = 0.19). Bootstrap analysis (n = 500) revealed substantial uncertainty in irrigated area estimates associated with threshold selection. Thus, the approach is best suited as a low-cost screening tool for identifying potential irrigated areas. In the Namoi catchment, such applications include irrigation hotspot detection, baseline mapping of irrigation extent and prioritisation of field scale irrigation surveillance. These findings demonstrate that the freely available satellite data can be used in place of the cost-and-labour-intensive field-based monitoring to identify potential irrigated areas using a simple transferrable method.