Scale Effects in Movable Bed Models of Rivers with Dominant Suspended Load

Movable bed physical models are often used as a tool for the prediction of the morphological behavior of rivers and estuaries. But as a tool they have limitations in their own in terms of so-called scale effects defined as differences between prototype and model due to deviations from some of the scale conditions. The consequences of these scale effects are sometimes serious as some important phenomena might not be reproduced correctly. This paper discusses for scale models of rivers in which suspended load is dominant the derivation of some important scale conditions and their application for the assessment of scale effects in reproducing flow, sediment transport and bed topography. Several numerical simulations were performed for a full scale U-shape prototype bend and five of virtual physical models based on three velocity and two bed material diameter scales using a depth-averaged 2D morphological model (Mike-21C). This allowed to quantify scale effects by comparing the prototype with the "virtual physical models" created in the mind only. The results show that (1) the selection of velocity scale is very important and (2) neither the Froude nor ideal velocity scale is a good choice for movable bed models for rivers with significant suspended load with respect to scale effects.