Effects of modified river cross-section geometry on hydraulic modelling performance: the slope break method

Sammanfattning: Remote sensing data has been widely used for large scale hydraulic modelling in the world. One of the most common ways to collect remote sensing data is by satellite. The digital elevation data observed from space is publicly accessible and covers almost all areas of the earth. However, the satellite cannot observe the river bathymetry. The satellite-derived river cross-sections only contain the bank information above the surface water level. The level of accuracy of river cross- sections derived from remote sensing data is not enough and it is possible to have a large influence on the model performance. This study will explore the capability of the slope-break method to improve the river bathymetry estimation and the influence of improved river cross-sections on the hydraulic model performance. First, four synthetic river cross-sections are constructed with consideration of the feature of satellite-obtained data. Second, the study tests the parameter influence in the slope-break method to improve the river bed elevation estimation. Third, the performance of hydraulic models is evaluated through several simulations, which focus on the input data collected during the different satellite observation period, the influence of modified submerged river cross-section shapes and the influence of two distinct flow conditions. The results show that the parameters of the slope-break method and satellite observation period have an inevitable impact on the accuracy of river bed elevation estimation, while the improved submerged shapes could compensate for the flaws of bed elevation estimation in the hydraulic modelling. The hydraulic models in this study are not sensitive to the flow conditions because of the simplicity of synthetic rivers. Nevertheless, the performance of hydraulic models with modified river cross-sections is more apparent in low flow conditions than in high flow conditions.