Home »
Academic Articles »
Applied Physics »
Integration of Hydrological Model and Time Series Model for Improving the Runoff Simulation: A Case Study on BTOP Model in Zhou River Basin, China
Integration of Hydrological Model and Time Series Model for Improving the Runoff Simulation: A Case Study on BTOP Model in Zhou River Basin, China
Qintai Xiao
(State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, No. 24 South Section 1, Yihuan Road, Chengdu 610065, China)
Li Zhou
(State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, No. 24 South Section 1, Yihuan Road, Chengdu 610065, China)
Xin Xiang
(State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, No. 24 South Section 1, Yihuan Road, Chengdu 610065, China)
Lingxue Liu
(Institute for Disaster Management and Reconstruction, Sichuan University-Hong Kong Polytechnic University, Chengdu 610065, China)
Xing Liu
(College of Water Resources and Hydropower, Sichuan Agricultural University, Ya’an 625014, China)
Xiaodong Li
(State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, No. 24 South Section 1, Yihuan Road, Chengdu 610065, China)
Tianqi Ao
(State Key Laboratory of Hydraulics and Mountain River Engineering, College of Water Resource and Hydropower, Sichuan University, No. 24 South Section 1, Yihuan Road, Chengdu 610065, China)
Journal: Applied Sciences
•
• MDPI AG
• ISSN: 2076-3417
Abstract
Improving the accuracy of runoff simulations is a significant focus of hydrological science for multiple purposes such as water resources management, flood and drought prediction, and water environment protection. However, the simulated runoff has limitations that cannot be eliminated. This paper proposes a method that integrates the hydrological and time series models to improve the reliability and accuracy of simulated runoffs. Specifically, the block-wise use of TOPMODEL (BTOP) is integrated with three time series models to improve the simulated runoff from a hydrological model of the Zhou River Basin, China. Unlike most previous research that has not addressed the influence of runoff patterns while correcting the runoff, this study manually adds the hydrologic cycle to the machine learning-based time series model. This also incorporates scenario-specific knowledge from the researcher’s area of expertise into the prediction model. The results show that the improved Prophet model proposed in this study, that is, by adjusting its holiday function to a flow function, significantly improved the Nash–Sutcliffe efficiency (NSE) of the simulated runoff by 53.47% (highest) and 23.93% (average). The autoregressive integrated moving average (ARIMA) model and long short-term memory (LSTM) improved the runoff but performed less well than the improved Prophet model. This paper presents an effective method to improve the runoff simulation by integrating the hydrological and time series models.
Qintai Xiao, Li Zhou, Xin Xiang, et al. (2022). "Integration of Hydrological Model and Time Series Model for Improving the Runoff Simulation: A Case Study on BTOP Model in Zhou River Basin, China". Applied Sciences. https://doi.org/10.3390/app12146883