长江科学院院报 ›› 2019, Vol. 36 ›› Issue (7): 23-27.DOI: 10.11988/ckyyb.20171318

• 水资源与环境 • 上一篇    下一篇

新安江流域abcd水量平衡模型及参数敏感性分析

吴光东1, 许继军1, Hoshin Gupta2, 张潇3   

  1. 1.长江科学院 水资源综合利用研究所,武汉 430010;
    2. 亚利桑那大学 水利和大气科学学院,美国 图森 85721;
    3. 长江勘测规划设计研究院 水利规划研究院,武汉 430010
  • 收稿日期:2017-11-15 出版日期:2019-07-01 发布日期:2019-07-18
  • 作者简介:吴光东(1989-),男,江苏徐州人,工程师,博士,主要从事水文水资源、水文模型的研究。E-mail:598105302@qq.com
  • 基金资助:
    国家重点研发计划项目(2016YFC0502201);长江科学院院级创新团队项目(CKSF2017061/SZ)

The “abcd” Water Balance Model: Application to Xin’an River Basin and Sensitivity Analysis

WU Guang-dong1, XU Ji-jun1, Hoshin Gupta2, ZHANG Xiao3   

  1. 1.Water Resources Department, Yangtze River Scientific Research Institute,Wuhan 430010, China;
    2.Department of Hydrology and Atmospheric Sciences, University of Arizona, Tucson 85721, USA;
    3.Water Resources Planning and Research Department, Changjiang Institute of Survey, Planning,Design and Research, Wuhan 430010, China
  • Received:2017-11-15 Online:2019-07-01 Published:2019-07-18

摘要: abcd水量平衡模型在国外得到了广泛应用,但在国内应用基本处于空白状态。为了探究该模型在我国中小流域应用的适应性及有效性,在介绍abcd模型的原理及结构的基础上,将其应用于新安江流域的径流预报模拟。在国内首次对abcd模型的4个参数进行敏感性分析,采用单变量法和多变量法分别探讨模型的性能指标对参数变化的响应程度。结果表明:校验阶段汛期与非汛期Nash确定性系数(Nash-Sutcliffe efficiency, NSE)分别为0.929和0.863,表明拟合程度较高;参数敏感性分析对于提高abcd模型参数率定的效率和精度具有显著意义;单变量法与多变量法结果相近,NSE对参数c最为敏感。通过对比径流模拟值与实测值,得出该模型模拟精度高、具有较高的适应性,能够广泛应用于我国中小流域。

关键词: abcd水量平衡模型, 径流模拟, 敏感性分析, 评价指标, 多变量法, 新安江流域

Abstract: Despite wide application abroad, the “abcd” water balance model is barely used in China. In order to explore the applicability and effectiveness of the “abcd” model in small and medium-sized watersheds in China, we first of all expound the principle and structure of the “abcd” model, and applied the model to the runoff forecasting simulation for Xin’an River Basin. Moreover, we examined the sensitivity of four parameters of the “abcd” model for the first time in China by comparing the simulated values with measured values, and further discussed the responses of performance indicators to parameter changes using univariate and multivariate methods. Results manifested that Nash-Sutcliffe efficiency (NSE)in verification is 0.929 in flood season and 0.863 in non-flood season, indicating a high fitting degree. Meanwhile, sensitivity analysis could evidently improve the efficiency and accuracy of calibration. Result of the univariate method is similar to that of the multivariate method, and NSE is the most sensitive to parameter c. In conclusion, the “abcd” model can be widely applied to small and medium-sized watersheds in China with high accuracy and adaptability.

Key words: abcd” water balance model, runoff simulation, sensitivity analysis, performance index, multivariate method, Xin’an River Basin

中图分类号: