Assessment of deficit irrigation impacts on water productivity and crop yield in the arid yellow river Basin using the SWAT model

Precision irrigation technologies, epitomized by drip irrigation, are crucial for mitigating agricultural water scarcity in arid regions. However, their widespread adoption is constrained by high initial investments, complex technical demands, and potential environmental risks. Consequently, a vast water-saving potential remains untapped within the less efficient, conventional irrigation systems that still dominate vast territories, such as Northwest China. This study presents a pragmatic and cost-effective pathway to enhance water productivity by evaluating alternative irrigation management. We developed and validated a coupled hydrological-crop growth framework using an improved Soil and Water Assessment Tool (SWAT) model for a large, canal-fed plain irrigation district in the Yellow River Basin. The model demonstrated high credibility, with a Nash-Sutcliffe efficiency NSE of 0.61 for runoff simulation and a coefficient of determination R² of 0.95 for maize yield estimation. The results reveal that strategic deficit irrigation, applied during non-critical growth stages, can reduce annual water diversion by 11 %–17 % while limiting yield losses to a marginal 2 %–3 %. This concurrently increases regional water productivity by 0.05–0.13 kg/m³ . Our study provides a model-driven framework demonstrating that achieving substantial water savings and ensuring stable food production is feasible without immediate, large-scale hardware investments.