Remote sensing-based assessment of long-term evapotranspiration and irrigation performance following modernization of Andalusian irrigation districts

Abstract This study develops a cloud-based remote sensing framework in Google Earth Engine to assess the evolution of evapotranspiration and irrigation performance across 49 representative irrigation districts in Andalusia (southern Spain), covering more than 212,000 ha during the period 2004–2019. The approach combines optical and thermal satellite data to estimate unstressed crop evapotranspiration ( ET us ) using vegetation-index-based crop coefficients and actual evapotranspiration ( ET a ) using the SSEBop energy balance model. These variables were integrated with irrigation adequacy indicators, including the Crop Water Stress Index ( CWSI ) and Relative Irrigation Supply ( RIS ), to evaluate long-term irrigation dynamics. Results reveal a clear divergence between ET us and ET a trends. ET us increased significantly across all irrigation typologies, driven by three concurrent processes: the progressive replacement of low-water-demand herbaceous crops, such as winter cereals, with woody perennials with expanding canopies; a shift within the remaining herbaceous fraction toward species with higher potential water requirements; and improved irrigation uniformity following the conversion of surface irrigation to pressurized systems. In contrast, ET a declined in inland irrigation systems and remained broadly stable in subtropical districts. This divergence indicates a sustained intensification of managed water stress, consistent with the widespread adoption of deficit irrigation strategies driven by rising irrigation costs, modernization of delivery systems, and regulatory constraints on water allocations. Regional analyses show no evidence of a rebound effect following modernization, as total actual crop water consumption slightly declined despite the increase in potential crop water demand.