Optimizing the Maize Irrigation Strategy and Yield Prediction under Future Climate Scenarios in the Yellow River Delta

Abstract

The contradiction between water demand and water supply in the Yellow River Delta restricts the corn yield in the region. It is of great significance to formulate reasonable irrigation strategies to alleviate regional water use and improve corn yield. Based on typical hydrological years (wet year, normal year, and dry year), this study used the coupling model of AquaCrop, the multi-objective genetic algorithm (NSGA-III), and TOPSIS-Entropy established using the Python language to solve the problem, with the objectives of achieving the minimum irrigation water (IW), maximum yield (Y), maximum irrigation water production rate (IWP), and maximum water use efficiency (WUE). TOPSIS-Entropy was then used to make decisions on the Pareto fronts, seeking the best irrigation decision under the multiple objectives. The results show the following: (1) The AquaCrop-OSPy model accurately simulated the maize growth process in the experimental area. The R2 values for canopy coverage (CC) in 2019, 2020, and 2021 were 0.87, 0.90, and 0.92, respectively, and the R2 values for the aboveground biomass (BIO) were 0.97, 0.96, and 0.96. (2) Compared with other irrigation treatments, the rainfall in the test area can meet the water demand of the maize growth period in wet years, and net irrigation can significantly reduce IW and increase Y, IWP, and WUE in normal and dry years. (3) Using LARS-WG (a widely employed stochastic weather generator in agricultural climate impact assessment) to generate future climate scenarios externally resulted in a higher CO2 concentration with increased production and slightly reduced IW demand. (4) Optimizing irrigation strategies is important for allowing decision makers to promote the sustainable utilization of water resources in the study region and increase maize crop yields.

Keywords

• AquaCrop-OSPy model
• maize
• NSGA-III
• TOPSIS-Entropy
• Yellow River Delta