The Dujiangyan Irrigation System (DIS), located in the western portion of the Chengdu Plain at the transitional junction between the Qinghai-Tibet Plateau and Sichuan Basin, has been in operation for about 2300 years. The system automatically uses natural topographical and hydrological features and provides automatic water diversion, sediment drainage and intake flow discharge control, thus preventing disastrous events in the region in a ‘natural’ way. Using a numerical modeling approach, this study aims to investigate the reasons behind this natural behavior of the system and provide a better understanding of the complex mechanisms which have caused the sustainability of the DIS for over two millennia. For this purpose, a two-phase flow model based on the Shallow Water Equations (SWEs) is developed to simulate the fluid and sediment motions in the DIS. A coupled explicit-implicit technique based on the Finite Element Method is applied for the fluid flow and a Sediment Mass (SM) model in the framework of the Lagrangian particle method is proposed to simulate the sediment motion under different flow discharge conditions. The results show how different components of the DIS make full use of the hydrodynamic and topographical characteristics of the river to effectively discharge sediment and excess flood to the downstream and create an environmentally sustainable irrigation system.

Compared with floods occurring over plains, alpine flash floods are formed over scattered locations with complex terrain and data is often lacking regarding the land topography, flow and sedimentation, causing difficulties when developing mathematical models to predict flash floods. The existing flash flood models mainly focus on the influence of water flow, such as the sharp increase in flow discharge caused by convergence at steep slopes, disregarding the sediment load carried by the water flow. However, under the effect of high-intensity sediment transport, the sedimentation in gullies may lead to surges in water level, causing the phenomenon of “great disasters resulting from minor flooding”. In this study, an efficient and accurate water-sediment coupling model was established with a Godunov-type finite volume method based 2D flow model, and a sediment module and OPENMP parallel computing module were added as well. Firstly, a common gully in mountainous area with large gradient variations was used as a generalized model to explore the impact of sedimentation on the flow field of the gully and compared with the physical model. Then, the alpine flash flood incident in Gengdi Village was simulated with the computer model. The calculation results show that the high-intensity sedimentation significantly increased the magnitude of alpine flash floods. Calculated by this mathematical model, the research results verified that this mathematic model can efficiently, accurately and concisely predict the occurrence of flash floods in gullies with large gradient variations. The model also provides a flow and sediment modeling method that incorporates the effect of high-intensity sediment transport into the traditional flash flood flow model. Thus, this model can be a powerful tool for detecting flash floods.

Xu Z, et al. Korean J Radiol. 2020 Feb;21:e55. https://doi.org/10.3348/kjr.2019.0853