Irrigation strategy optimization in irrigation districts with seasonal agricultural drought in southwest China: A copula-based stochastic multiobjective approach

Seasonal agricultural droughts (SAD) often cause yield loss for the farming community around the world, reinforcing the need to optimize both water-saving measures and irrigation schedules to improve irrigation water use efficiency. This study proposed a Copula-based stochastic multiobjective programming (C-SMP) model for optimizing irrigation strategies to mitigate the negative impacts of SAD. The model uses the regional net irri-gation water demand and stream runoff to formulate the Copula, which better characterizes SAD. Drought scenarios are selected based on reservoir storage capacity and irrigation water use efficiency. Then, randomness of drought scenarios from Copula model, multiple conflicting objectives, selecting of water-saving measures, and deficit irrigation strategy optimization were addressed via the stochastic multiobjective programming technique. A case study was conducted in the Dongfeng reservoir irrigation district (DFRID), Meishan City of Sichuan Province, southwest China to optimize both water-saving measures and deficit irrigation schedules for different crops during their growth periods. The results showed that: (1) the joint distribution model established by the Frank Copula function had the best performance in measuring SAD, and the probability of drought occurrence was 25% in DFRID; (2) optimal irrigation strategies improved the yield of wheat and citrus by around 12.51-32.22% and 8.33-12.83%, respectively, compared with the average level of Meishan City. Only the yield of rape was reduced by - 27.68 similar to-1.42% under the extreme dry condition of net irrigation water demand, while it increased by 1.84-4.56% when the net irrigation water demand (NIWD) was in a dry condition; (3) the C-SMP demonstrated clear advantages in dealing with multiple conflicting objectives and the randomness of SAD sce-narios. The proposed approach is applicable for irrigation districts suffering from SAD and can help formulate more effective and sustainable irrigation strategies.

Automated summary

This study proposed a Copula-based stochastic multiobjective programming model for optimizing irrigation strategies to mitigate the negative impacts of seasonal agricultural droughts (SAD). The model uses regional net irrigation water demand and stream runoff to better characterize SAD. A case study in Meishan City of southwest China showed that the optimal irrigation strategies improved crop yields and the C-SMP demonstrated clear advantages in dealing with multiple conflicting objectives and the randomness of SAD scenarios.