Scale effects in 3-D modelling of straight compound channels

Abstract

Current understanding of compound channel behaviour is heavily based on findings from small scale experimental flumes. Little work has been done to date to assess whether the turbulent momentum exchange associated with velocity gradients between the main channel and floodplain zones of these channels is significant at larger scales, typical of natural channels. This paper assesses the performance of the Divided Channel Method (DCM) which is still commonly used by the engineering community for analysing compound channel flows, and the 3-Dimensional hydrodynamic codes SSIIM and TELEMAC-3D. Calculated and predicted results are compared to results measured from the small scale laboratory flume at the University of Ulster, the large-scale UK Flood Channel Facility and the River Main in Northern Ireland.