Compound Channel Modelling Capability Assessment of 1D Hydraulic Model

Accurate modelling of streams or open channels in urban catchment is vital for different types of engineering and environmental studies such as flood risk assessments, flood alleviation designs, water quality and ecological assessments.

The estimation of conveyance in existing one-dimensional hydrodynamic modelling software such as MIKE URBAN, MIKE11 and HECRAS is principally based on some form of the Manning Equation. This is an empirically derived equation and does not provide meaningful results for sudden changes in wetted perimeter with depth as generally encountered in natural watercourses comprising of compound sections with significant variations in flow velocity across the cross-section. A number of methods are used in existing one-dimensional hydrodynamic modelling packages to circumvent this physically incorrect results. Computation of conveyance in compound channel sections is somewhat different in various modelling packages resulting different flood level predictions.

This paper presents a modelling exercise for benchmarking the compound channel modelling capability of 1D hydraulic model (MIKE URBAN/MIKE11) by comparing modelling results with that of 2D (MIKE21) and 1D-2D (MIKE11-MIKE21) coupled modelling packages for Oteha Stream catchment under identical network, topographical and boundary conditions.