长江科学院院报 ›› 2025, Vol. 42 ›› Issue (3): 202-210.DOI: 10.11988/ckyyb.20241180

• 长江源科学考察与研究专栏 • 上一篇    

基于嵌套主成分回归模型的长江源区径流重建

姜晓萱1,2(), 王文卓3(), 袁喆1,2,4, 霍军军1,2,4, 周涛1,2,4   

  1. 1 长江科学院 流域水资源与生态环境科学湖北省重点实验室,武汉 430010
    2 长江水利委员会长江经济带保护与发展战略研究中心,武汉 430010
    3 河海大学 水文水资源学院,南京 210003
    4 长江科学院 水资源综合利用研究所,武汉 430010
  • 收稿日期:2024-11-18 修回日期:2025-01-30 出版日期:2025-03-14 发布日期:2025-03-14
  • 通信作者:
    王文卓(1993-),女,河南郑州人,副研究员,博士,主要从事径流重建与预测的相关研究。E-mail:
  • 作者简介:

    姜晓萱(2001-),女,湖北荆门人,硕士研究生,主要从事流域水资源规划管理的相关研究。E-mail:

  • 基金资助:
    国家重点研发计划项目(2023YFC3206602)

Reconstruction of Runoff in the Source Region of Yangtze River Based on Nested Principal Component Regression Modeling

JIANG Xiao-xuan1,2(), WANG Wen-zhuo3(), YUAN Zhe1,2,4, HUO Jun-jun1,2,4, ZHOU Tao1,2,4   

  1. 1 Hubei Provincial Key Laboratory of Basin Water Resources and Eco-environmental Sciences,Changjiang River Scientific Research Institute,Wuhan 430010,China
    2 Research Center on Protection and Development Strategy for Yangtze River Economic Belt, Changjiang Water Resources Commission, Wuhan 430010, China
    3 College of Hydrology and Water Resources, Hohai University, Nanjing 210003, China
    4 Water Resources Department, Changjiang River Scientific Research Institute, Wuhan 430010, China
  • Received:2024-11-18 Revised:2025-01-30 Published:2025-03-14 Online:2025-03-14

摘要:

基于嵌套主成分回归模型,重建了长江源区公元1433—2002年的天然年径流序列,探讨了其历史变化特征与干湿事件演变及周期性波动。利用树轮数据及径流观测数据,结合模型评估指标(CRSQ、VRSQ、RE、CE),证明了该模型的重建精确性。研究结果表明,在过去570 a中,长江源区年径流量整体呈显著波动,识别出6个湿润期与9个干旱期,其中1451—1510年和1596—1645年是最长的湿润期,1848—1903年是持续时间最长的干旱期。重建时期的干旱与长江流域青藏高原等地区的干旱时期较为吻合,表明长江源区重建径流变化能够反映大尺度的气候波动。此外,长江源区重建径流序列存在4~8、16~32、50~100、100~200 a的显著周期性波动,可能受ENSO(厄尔尼诺-南方涛动)、东亚夏季风(EASM)变化、太平洋年代际振荡(PDO)及大西洋多年代际振荡(AMO)共同驱动,也反映了全球气候长期变化及青藏高原冰川与积雪消融趋势的影响。

关键词: 径流量, 重建, 树木年轮, 嵌套主成分回归模型, 长江源区

Abstract:

This study employs a nested principal component regression model to reconstruct the natural annual runoff series for the Yangtze River source region from 1433 to 2002. It explores historical variability, the evolution of wet and dry events, and periodic fluctuations. Using tree-ring data and observed runoff data, the model’s accuracy is validated through evaluation indicators (CRSQ, VRSQ, RE, and CE). Results indicate that over the past 570 years, the annual runoff in the Yangtze River source region has experienced significant fluctuations. Six wet periods and nine dry periods are identified, with the longest wet periods occurring from 1451 to 1510 and from 1596 to 1645, and the longest dry period spanning from 1848 to 1903. The dry periods during the reconstruction period align with droughts on the Tibetan Plateau and in other areas of the Yangtze River basin, suggesting that the reconstructed runoff changes in the Yangtze River source region reflect large-scale climate fluctuations. Furthermore, the reconstructed runoff series for the Yangtze River source region exhibits significant periodic fluctuations at intervals of 4-8 years, 16-32 years, 50-100 years, and 100-200 years. These fluctuations are likely driven by the combined effects of ENSO (El Niño-Southern Oscillation), the East Asian Summer Monsoon (EASM), the Pacific Decadal Oscillation (PDO), and the Atlantic Multidecadal Oscillation (AMO), and also reflect the impact of global climate change and the trends of glacier and snowmelt on the Tibetan Plateau.

Key words: runoff, reconstruction, tree annual rings, nested principal component regression model, Yangtze River source region

中图分类号: