Zhou Yan, Xu Xianghui, Li Mo, Zhang Xinrui, Cao Kaihua
School of Water Conservancy & Civil Engineering, Northeast Agricultural University, Harbin, Heilongjiang, 150030, China.
College of Engineering, Northeast Agricultural University, Harbin, Heilongjiang, 150030, China.
J Environ Manage. 2022 Jul 1;313:114945. doi: 10.1016/j.jenvman.2022.114945. Epub 2022 Mar 31.
The uncertainty of the hydrological environment and unbalanced water resource allocation result in a high risk of irrigation water shortages in regional agriculture, which seriously affects the sustainable development of agricultural systems. In this paper, we propose a risk regulation based modeling approach for the optimal allocation of agricultural water resources in a complex stochastic environment. The approach includes a conditional value-at-risk (CVaR) model, two-stage stochastic programming (TSP) model, two-dimensional joint distribution probability (JP) model, fractal criteria, and a multiple forms of chance-constrained programming (CCP) model. The model can weigh the contradiction between the intended target and associated penalties attributed to unknown hydrological events, measure the risk between system benefits and expected losses in agricultural water allocation at different confidence levels, and address the randomness in the objective function and constraints (including the left end term, right end term, and left and right end terms). To verify the applicability of the method, it is applied to the Jinxi Irrigation District in China to optimize the allocation and risk regulation of limited water resources under the variable runoff conditions of the Songhua River and crop water demands in the irrigation area. By adjusting parameters such as risk preference and probability of violation, the risk of water shortages in the irrigation area can be regulated, and the multidimensional impacts of different water allocation schemes on agricultural economic benefits, social benefits, ecology and environment can be determined. The case study reveals that the CTSP-CCJP method is sensitive, applicable to complex and uncertain environments and important for the efficient use of agricultural water resources and risk reduction.
水文环境的不确定性和水资源分配不均衡导致区域农业灌溉缺水风险高，严重影响农业系统的可持续发展。本文提出一种基于风险调控的建模方法，用于复杂随机环境下农业水资源的优化配置。该方法包括条件风险价值（CVaR）模型、两阶段随机规划（TSP）模型、二维联合分布概率（JP）模型、分形准则以及多种形式的机会约束规划（CCP）模型。该模型能够权衡预期目标与未知水文事件所致相关惩罚之间的矛盾，度量不同置信水平下农业水资源配置中系统效益与预期损失之间的风险，并处理目标函数和约束条件中的随机性（包括左端项、右端项以及左右端项）。为验证该方法的适用性，将其应用于中国的锦西灌区，以优化松花江径流变化条件及灌区作物需水情况下有限水资源的配置与风险调控。通过调整风险偏好和违约概率等参数，可调控灌区缺水风险，并确定不同水资源配置方案对农业经济效益、社会效益、生态与环境的多维度影响。案例研究表明，CTSP - CCJP方法灵敏，适用于复杂不确定环境，对农业水资源高效利用和降低风险具有重要意义。
