Large-scale morphology, in particular meander bend depth, bar dimensions, and bifurcation dynamics, are greatly affected by the deflection of sediment transport on transverse bed slopes due to gravity and by secondary flows. Overestimating the transverse bed slope effect in morphodynamic models leads to flattening of the morphology, while underestimating leads to unrealistically steep bars and banks and a higher braiding index downstream. However, existing transverse bed slope predictors are based on a small set of experiments with a minor range of flow conditions and sediment sizes, and in practice models are calibrated on measured morphology. The objective of this research is to experimentally quantify the transverse bed slope effect for a large range of near-bed flow conditions with varying secondary flow intensity, sediment sizes (0.17–4 mm), sediment transport mode, and bed state to test existing predictors. We conducted over 200 experiments in a rotating annular flume with counterrotating floor, which allows control of the secondary flow intensity separate from the streamwise flow velocity. Flow velocity vectors were determined with a calibrated analytical model accounting for rough bed conditions. We isolated separate effects of all important parameters on the transverse slope. Resulting equilibrium transverse slopes show a clear trend with varying sediment mobilities and secondary flow intensities that deviate from known predictors depending on Shields number, and strongly depend on bed state and sediment transport mode. Fitted functions are provided for application in morphodynamic modeling.

@en

We experimentally tested the effect of a  largerange in helical flow intensity and  sedimentmobility on equilibrium transverse  slopes,which resulted in a function for slope  effectsdepending on bed state and  sedimenttransport mode, that deviates from  linearfunctions with sediment mobility in  literature.Furthermore, we obtained basic  relationsfor sorting patterns as a function of  transverseslope and sediment mobility 

@en

Lateral sorting (= bend sorting) is observed in  natural meanders, where the inner and outer  bend are fairly fine and coarse, respectively  (e.g. Julien and Anthony, 2002; Clayton and  Pitlick, 2007). This is caused by the mass  differences between grains on a transverse  slope, leading to coarser grains being dragged  down net more than finer grains (Ikeda et al.,  1987). The slope of the transverse bed  influences the degree of bend sorting greatly.  Also vertical sorting occurs. Grainflows at  the lee side of dunes result in a net fining  upward trend (Kleinhans, 2005). The degree of  vertical sorting depends on flow velocity and  sediment mixture characteristics.  Most previous studies focused on the  development of the transverse slope using  small ranges of uniform sediment, so spatial  sorting was absent. Yet, it was argued that  there is a feedback mechanism between bend  sorting and the transverse slope (Ikeda et al.,  1987). So, it is of key importance to attain  better understanding of how sorting in river  bends comes about, which can improve current  numerical models.  It is the objective of this study to examine  experimentally the effect of transverse bed  slope and sediment mobility on spatial sorting  of bed load in a meander. 

@en

We experimentally tested the effect of a large range in helical flow intensity and sediment mobility on equilibrium transverse slopes, which resulted in a function for slope effects depending on bed state and sediment transport mode, that deviates from linear functions with sediment mobility in literature. Furthermore, we obtained basic relations for sorting patterns as a function of transverse slope and sediment mobility.@en