考虑降水‑径流‑需水丰枯遭遇的宝鸡峡灌区水库供水风险分析

Risk Analysis of Reservoir Water Supply in Baojixia Irrigation Area Considering Wetness‑Dryness Encounter of Precipitation, Runoff and Water Demand

摘要:

气候变化和高强度人类活动显著增加了水文过程和用水过程的不确定性，给灌区水资源调控带来风险。针对非同频率降水量、径流量、需水量三变量丰枯遭遇组合不确定性导致的灌区水库供水风险等问题，本文提出“供需水联合概率分析?水库优化调度?供水风险识别”研究框架。考虑灌区降水、需水与河源来水的非线性相依关系，构建Vine Copula联合概率分布模型分析供需水丰枯遭遇组合概率;构建权衡公平与效益、农业与生态用水等多目标的水库群优化调度模型，提取不同供?需水丰枯遭遇情景下水库调度方案;耦合多属性决策云模型和水库群供水风险评估指标体系，分析不同调度方案的供水风险。将提出的研究框架应用于宝鸡峡灌区水库调度，结果表明：Vine Copula可有效表征非同频率降水量?径流量?需水量三变量的联合不确定性，其中，情景DDW（降水枯、径流枯、需水丰）联合概率（13.56%）最大，与实际遭遇概率（13.33%）接近;多目标水库调度模型能有效权衡公平与效益、农业用水与生态用水间的竞争关系，不同调度方案基尼系数均小于0.4，灌溉缺水率小于42.0%，经济效益为7.61亿~8.12亿元，生态缺水量小于1.17×10^8 m3;调度方案C4所对应的情景DDW（降水枯、径流枯、需水丰）供水风险最高，为49.02%，综合风险等级为Ⅲ级。研究结果可为灌区供需水联合不确定性的水库群供水风险提供科学依据。

Abstract:

Climate change and high?intensity human activities have significantly increased the uncertainty of hydrological and water usage processes, posing risks to water resource regulation in irrigation districts. In response to the uncertainties caused by the combination of different frequency precipitation, runoff, and water demand, a research framework of “joint probability analysis of supply and demand?optimal scheduling of reservoirs?identification of water supply risks” was proposed. Considering the nonlinear interdependence among precipitation, water demand, and river source inflow in the irrigation district, a Vine Copula joint probability distribution model was proposed to analyze the probability of the combination of water supply and demand under different wet and dry conditions. A multi?objective reservoir group optimal scheduling model was constructed to balance fairness and efficiency, as well as the competition between agricultural and ecological water use, and to extract reservoir scheduling schemes under different supply and demand wet and dry conditions. A multi?attribute decision?making cloud model and a reservoir group water supply risk assessment index system were coupled to analyze the water supply risks of different scheduling schemes. The proposed research framework was applied to the reservoir scheduling of the Baojixia Irrigation Area. The results showed that Vine Copula can effectively represent the joint uncertainty of the three variables of precipitation, runoff, and water demand with different frequencies. Among them, the joint probability of scenario dry precipitation, dry runoff, and abundant water demand (DDW) was the highest (13.56%), which was close to the actual encounter probability (13.33%). The multi?objective reservoir scheduling model can effectively balance the competition between fairness and efficiency, as well as agricultural and ecological water use. The Gini coefficient of different scheduling schemes was all below 0.4, the irrigation water shortage rate was less than 42.0%, the economic benefit was 761 million~812 million yuan, and the ecological water shortage was less than 1.17×10^8 m3. The scheduling scheme C4 corresponding to scenario dry precipitation, dry runoff, and abundant water demand (DDW) had the highest water supply risk, which was 49.02%, and the comprehensive risk level was grade Ⅲ. The research can provide a scientific basis for the water supply risk of reservoir groups under the joint uncertainty of water supply and demand in irrigation districts.