太湖流域上游不同功能性水库流域径流响应模拟与预测
DOI:
作者:
作者单位:

1.江苏第二师范学院地理科学学院;2.河海大学水文与水资源学院;3.中国科学南京地理与湖泊研究所;4.郑州市明新中学

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金(42530511、42371043);国家重点研发计划项目(2023YFC3208704)


Simulation and Prediction of Hydrological Responses in Different Functional Reservoir Basins in the Upper Reaches of the Taihu Lake Basin
Author:
Affiliation:

Fund Project:

National Natural Science Foundation of China (No. 42530511, No. 42371043);National Key Research and Development Program of China (No. 2023YFC3208704)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    [目的]揭示全球变暖和强人类活动影响下太湖流域水文响应变化的空间异质性与气候敏感性特征,量化太湖流域上游不同功能水库流域的径流变化规律及驱动机制,为保障未来长三角地区防洪供水安全与生态安全提供科学支撑。[方法]以太湖流域上游不同功能定位下的沙河水库与横山水库流域为研究对象,构建基于长短期记忆网络(LSTM)和贝叶斯自动优化算法的水文响应模拟与预测框架,结合历史(2006—2021年)降雨、流量实测数据和CMIP6未来SSP2-4.5情景下降水数据(2022—2037年、2038—2053年),量化分析不同功能性水库流域径流变化趋势及原因。[结果]LSTM模型在日流量和洪水事件模拟中表现优异,沙河水库流域洪量、洪峰、峰现时间合格率均达100%,横山水库流域洪峰与峰现时间合格率为100%,洪量合格率为67%。在降水变化与水库功能性调度交互影响下,2022—2037年、2038—2053年沙河水库流域日平均流量呈现先减少(由2.51 m3/s下降到1.18 m3/s)后增加趋势(从1.18 m3/s增至4.66 m3/s),极端高流量事件频次与强度显著上升;横山水库流域日平均流量则持续递减(从3.14 m3/s降至2.29 m3/s,再减少至0.83 m3/s),极端洪水事件风险降低。[结论]建议未来沙河水库需强化高流量防洪调控体系建设,横山水库则需侧重于低流量生态供水安全保障能力提升。研究成果为气候变化背景下丘陵型水库流域水文响应模拟及差异化应对策略制定提供了科学方法和依据。

    Abstract:

    [Objective]The characteristics of spatial heterogeneity and climate sensitivity of hydrological response changes in the Taihu Lake Basin under the influence of global warming and intense human activities were analyzed, the laws and driving mechanisms of runoff changes in the basins of reservoirs with different functions in the upper reaches of the Taihu Lake Basin were quantified, and a scientific basis was provided for ensuring the future flood control, water supply security, and ecological security in the Yangtze River Delta region.[Methods]The Shahe Reservoir Basin and Hengshan Reservoir Basin with different functional orientations in the upper reaches of the Taihu Lake Basin were taken as the research objects. A hydrological response simulation and prediction framework based on the Long Short-Term Memory (LSTM) network and Bayesian automatic optimization algorithm was constructed. Combined with historical (2006—2021) observed rainfall and discharge data, as well as future precipitation data under the CMIP6 SSP2-4.5 scenario (2022—2037, 2038—2053), the trends and causes of runoff changes in the basins of reservoirs with different functions were quantitatively analyzed.[Results]The LSTM model exhibited excellent performance in simulating daily discharge and flood events. For the Shahe Reservoir Basin, the qualification rates of flood volume, flood peak, and peak occurrence time all reached 100%; for the Hengshan Reservoir Basin, the qualification rates of flood peak and peak occurrence time were 100%, while that of flood volume was 67%. Under the interactive influence of precipitation changes and functional reservoir operation, the daily average discharge in the Shahe Reservoir Basin showed a trend of first decreasing (from 2.51 m3/s to 1.18 m3/s) and then increasing (from 1.18 m3/s to 4.66 m3/s) during 2022—2037 and 2038—2053, with a significant increase in the frequency and intensity of extreme high-flow events. In contrast, the daily average discharge in the Hengshan Reservoir Basin continued to decrease (from 3.14 m3/s to 2.29 m3/s, then to 0.83 m3/s), and the risk of extreme flood events was reduced.[Conclusion]Strengthening the construction of high-flow flood control and regulation systems for the Shahe Reservoir in the future, and focusing on improving the capacity to ensure the safety of ecological water supply under low-flow conditions for the Hengshan Reservoir, will be essential for addressing the hydrological response challenges of the two reservoirs. The research results provide scientific methods and basis for simulating hydrological responses in hilly reservoir basins and formulating differentiated response strategies under the background of climate change.

    参考文献
    相似文献
    引证文献
引用本文
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2025-10-21
  • 最后修改日期:2026-05-16
  • 录用日期:2026-05-19
  • 在线发布日期:
  • 出版日期: