Application of The Independent Subsections Method for the Estimation of the Rating Curve in the Compound Channel – A Case Study

The exchange discharge and the independent subsections methods are two state-of-the art methods for estimation of a stage-discharge curve in a compound channel and 1D non-uniform flow computations.Traditional, divided channel method and the exchange discharge method,are based on the following three assumptions: 1) the dynamic equation for the water surface profile computations is written for the total cross-section, 2) the head loss gradients in the subsections are equal in all subsections and 3) uniform flow conditions are prescribed for the downstream discharge distribution. These assumptions cannot provide satisfactory prediction of both flow depths and the discharge distribution between subsections. Proust et al. 2009 tried to overcome this drawback by extending the method proposed by Yen et al. 1985 who divided the compound cross-section into subsections according to changes in geometry and roughness and derived a set of 1D flow equations by writing the mass and momentum equation for each subsection.The extension included special treatment of momentum exchange at the interface of adjacent subsections, thus allowing computation in non-prismatic straight compound channels with constant or variable total channel width. The method was named independent subsections method (ISM). It was successfully tested against the experimental data for uniform and non-uniform flows in two-stage laboratory canals with smooth and rough floodplains, as it was done with the exchange discharge model. However, the existing set of equations is derived for simplified compound channel geometry, i.e. the one which consists of rectangular subsections. Thus, the aim of this paper is to further extend the model to arbitrary geometries and to validate the method against the available floods observation data from one gauging station in Serbia. Momentum equations will also account for the effect of vegetation on floodplains by introducing additional term that describes the influence of the volume drag force exerted by an array of rigid vegetation stems as described by Proust et al. 2016. Results will be discussed in terms of discharge distributions, floodplain flow depths and contributions of different sources of energy dissipation to the total head loss, i.e. 1) the bed friction, 2) the turbulent momentum flux,3) the momentum flux due to mass exchange and 4) the volume drag force exerted by rigid-stem vegetation.