Nicolas Huybrechts

Dynamic Routing of Flow Resistance and Alluvial Bed-Form Changes from the Lower to the Upper Regime

On the basis of a Strouhal number and the definition of the control factor, m, a new routing to calculate the energy slope in the lower and upper alluvial regimes is proposed. The control factor, m, representing the interactions in alluvial rivers, is reckoned as a bed-form index: while the flow evolves through transition, the control factor, m, decreases from m=2, associated with two-dimensional fully developed dunes, to m=1, associated primarily with in-phase waves. The way to predict the value of the control factor, m, is drawn from a previously published criterion for delineating the upper regime and is calibrated with experimental data. On several data from flumes and rivers, the routing is tested and compared with other methods from the literature. It appears that the new routing is the most robust because it allows researchers to obtain low averages of the discrepancy ratio for a wide range of ratios between the water depth and the median sediment diameter. On a selection of contrasted freshet events, the new routing allows for the capture of the primary dynamic of the flow resistance decrease.

Observations of canonical flow resistance in fast-flowing sand bed rivers

Field data are here re-interpreted using the Vortex–Drag equation for evaluating the alluvial resistance in a sand bed river. The Vortex–Drag equation introduces the Rossiter resonance concept into alluvial hydraulics. It appears that this equation is more consistent than the Manning approach for the lower alluvial regime since a ripple configuration effectively induces a lower flow resistance coefficient than a duned configuration. A systematic decrease in the control factor down to a canonical value during the transition from the lower to the upper alluvial regime is found for well-contrasted freshet events. It results that during high stream power events with maximum river bed reshaping, the flow pattern is reproduced by simply imposing this canonical value. This is most welcome in alluvial hydraulic routing, because peak-discharge events are generally these which are the most difficult to characterize in a straightforward way, both experimentally and numerically.

A new closure methodology for 1D fully coupled models of mobile-bed alluvial hydraulics: Application to silt transport in the Lower Yellow River

Abstract A new closure approach involving a common parameter has been incorporated into a 1D fully coupled model of mobile-bed alluvial hydraulics. The objective is to simplify the methodology of 1D river routing models and to improve their accuracy. The common parameter, called control factor m , introduces the concept of Rossiter modes in alluvial hydraulics and represents the interactions between the flow, the sediment transport and the bed morphology. The feasibility of the new closure approach has been established by reproducing numerically the 2002 silt flushing experiment conducted on the Lower Yellow River (LYR) downstream the Xiaolangdi reservoir. From the comparison between the experimental data and the numerical results, a time evolution of the control factor m reproducing the characteristics of the flow has been extracted. This time evolution agrees with analysis conducted previously on other datasets and with data measured during the flush. The results obtained with this time evolution for the hydraulics, the sediment transport and bed adaptation are encouraging but still need improvements and further feeding from complementary experimental data.

Coupled estimation of the energy slope and associated sand-silt transport during high stream power events in alluvial rivers

Abstract A coupled routing for the transport capacity and the energy slope is introduced through the definition of the control factor m whose value is linked to the bed form configuration. The coupling aims to further incorporate the interactions occurring in alluvial rivers and thus enhance the prediction of the fine sediment fluxes, especially during high stream power events. Based on a predictive rule for the control factor m that only involves water depth, velocity and bedform constitutive texture, the novel method is confronted to observations collected in one of the most strongly dynamic alluvial river namely the Lower Yellow River. Comparisons between time series of measured and computed concentrations illustrate that during high velocity events the main dynamics of the sediment transport is correctly reproduced. The main advantage of the present approach is to supply consistent time evolutions of sediment concentrations without making use of any detailed shear information.