| This article has been：Browse 346Times Download 533Times |
scan it! |
|
|
| DOI：10.13522/j.cnki.ggps.2024425 |
|
| Runoff variation across multiple time scales and their future changes in the Gansu Section of the Weihe River Basin |
|
WANG Yiru, SUN Dongyuan, WANG Xingfan, NIU Zuirong, CUI Yanqiang, MA Yali, WU Lanzhen, SHU Heping
|
|
College of Water Conservancy and Hydropower Engineering, Gansu Agricultural University, Lanzhou 730070, China
|
| Abstract: |
| 【Objective】 The Gansu Section of the Weihe River Basin plays a critical role in regional water supply and ecological stability. This paper investigates the multi-time scale variations of runoff and their potential changes in the future, aiming to provide a basis for improving management and sustainable use of water resources in the basin.【Method】The analysis was based on monthly runoff data measured from 1956 to 2022 from the hydrological stations at Wushan and Beidao. The Mann-Kendall trend test, the rescaled range (R/S) analysis, along with other statistical methods, were used to analyze intra-annual, inter-annual, and seasonal runoff variations. The CNN-LSTM- Attention hybrid deep learning model was developed to predict future runoff changes.【Result】The annual runoff in the Gansu Section of the river basin showed pronounced seasonal variation, with the majority occurring from July to October. Although the annual runoff distribution was uneven, a gradual shift toward more uniformity was observed from 1956 to 2022. Both annual and seasonal runoff showed a declining trend from 1956 to 2022, with a significant abrupt change observed in the 1980s, after which runoff volumes dropped notably. Multi-scale analysis revealed alternating periods of high and low runoff. Model forecasts suggest that the runoff will continue to decline from 2023 to 2030.【Conclusion】Runoff variability in the Gansu Section of the Weihe River is driven by both climate change and human activities, imposing a challenging outlook for sustainable use of water resources in the region. |
| Key words: runoff evolution; runoff prediction; Gansu Section of Weihe river; CNN-LSTM-Attention |
|
|
