Spatio-Temporal Optimization of Rice Irrigation at Raster Scale: Synergies Between Water Productivity and Methane Emission Reduction

This study addresses the challenges of coordinating spatio-temporal water allocation to optimize water productivity and reduce carbon emissions in water resource management, particularly the lack of high-resolution, integrated optimization frameworks capable of simultaneously tackling water scarcity and greenhouse gas (GHG) emissions. We propose a modeling approach for large-scale regional rice irrigation that explicitly represents the physical-process-based relationships among irrigation water, yield, and methane (CH4) emissions. Using GIS, a grid-based simulation domain was constructed at a 500 m × 500 m resolution, and the GIS-DSSAT and GIS-DNDC models were employed to simulate yield and CH4 emissions under varying irrigation amounts. The Random Forest algorithm—selected for its ability to capture complex nonlinear interactions—was used to establish the response surfaces linking irrigation water, yield, and CH4 emissions. A spatio-temporal irrigation optimization model was then developed to simultaneously reduce CH4 emissions and enhance water productivity. This methodology was applied to the Sanjiang Plain in Heilongjiang Province, where the NSGA-II algorithm was used to derive optimal irrigation schemes for rice cultivation across 408,264 grid cells. The results revealed quadratic nonlinear relationships between irrigation water amount, yield, and CH4 emissions. Compared to the conventional irrigation practice in the region, which typically involves 15–20 flood irrigation events per season, the optimized irrigation schedule comprised 7–14 events—with 12 events accounting for 42% of the cases—and an irrigation duration ranging from day 137 to 256. This led to a 10.3% reduction in total irrigation volume, a 9.6% decrease in CH4 emissions per unit yield, and a 21.8% increase in water productivity. This study provides valuable decision support for optimizing regional water allocation and developing rice cultivation strategies that improve productivity while reducing emissions.