Author(s): S. Atabay; D.W. Knight
Keywords: Boundary shear; conveyance; floods; flume experiments; overbank flow; sediment transport
Abstract: Several sets of experimental data on conveyance capacity, the division of flow between a main river channel and its floodplains, boundary shear, and sediment transport in overbank flow are presented. The strengths and weaknesses of various 1-D modelling approaches are highlighted. The coherence method (COHM) of Ackers [J. Hydraul. Res. 31 (1993) 509] and the weighted divided channel method (WDCM) of Lambert and Myers [Proc. Inst. Civil Engineers Water Maritime Energy 130 (1998) 84] are shown to be two useful 1-D methods for dealing with overbank flows. They are relatively simple to apply, and give not only the stage–discharge relationship, but also the division of flow between the main channel and the floodplains. Their general validity has been extended by testing the two methods against data sets other than those used in their original formulation.A modified version of the WDCM is presented that gives better predictions for homogeneously roughened channels. It is shown that the WDCM does not give good predictions for heterogeneously roughened compound channels. A comparison is made between the COHM and WDCM concerning their relative accuracy and ease of application for conveyance and boundary shear calculations. Finally, the COHM is applied to some mobile boundary compound channel data and shows that once the total and zonal channel discharges are calculated correctly, the sediment transport concentration may be predicted reasonably well for high discharges but not so well for low discharges and shallow floodplain depths, unless the boundary shear stress on the main channel bed is specified or calculated.