Prediction of movable bed resistance in open channels with dune forms incorporating the effects of flow separation and water temperature

Abstract

Flow separation behind the dune forms may have reduced the contact area between the flows and the bed surface, in addition to producing form drag in open channels. In this study, a novel predictor is developed based on the underlying mechanism of movable bed resistance across a dune-dominated bed. The mechanism is derived to a new formula based on the energy loss division approach to generalize the relationships among the grain drag, form drag and flow separation. The flow separation length is expressed to quantify the flow separation effect with a modified empirical method. Additionally, the water temperature effect is incorporated into the novel predictor due to the inclusion of fluid viscosity in both the roughness function and suspension number. A global database with a wide spectrum of measurements from the available literature was created to test the novel predictor. Approximately 83.3 % of flume data fall within a range of ± 30 % error, and ∼ 80.0 % of field data plot within this range. The optimized predictor has a better accuracy than the other four previously developed models in all five evaluation indicators. Furthermore, this predictor effectively reflects the impacts of water temperature on flow resistance to reproduce four variation modes. The diverse modes are jointly determined by suspension number and roughness Reynolds number. In actual river management, this predictor is used to interpret the effects of lower water temperature on sediment carrying capacity and flood routing. Seasonal variations in water temperature and the underlying effects of water temperature occur in rivers worldwide, which implies the need to pay more attention to this aspect in river management strategies.