Estimation of bathymetry (and discharge) in natural river cross-sections by using an entropy approach

This paper presents a new method for reconstructing the bathymetric profile of a cross section based on the application of the principle of maximum entropy and proposes a procedure for its parameterization. The method can be used to characterize the bathymetry of a cross-section based on a reduced amount of data exclusively of a geometric type, namely, the elevation of the lowest point of the channel crosssection, the observed, georeferenced flow widths and the corresponding water levels measured during the events. The procedure was parameterized and applied on two actual river cross-sections characterized by different shapes and sizes. In both cases the procedure enabled us to describe the real bathymetry of the cross-sections with reasonable precision and to obtain an accurate estimate of the flow areas. With reference to the same two cases, we show, finally, that combining the bathymetry reconstruction method proposed here and an entropy-based approach for estimating the cross-sectional mean velocity previously proposed (Farina et al., 2014) enables a good estimate of discharge.