Abstract:
Compound channels are a common configuration of rivers. During extreme events of
floods, the momentum transfer due to the difference of the velocities between the main
channel and the floodplains flows generates a complex 3D flow. Accurate estimation of
channel capacity remains a difficult issue.
Although several studies have being carried out in the past, a new experimental study
aims revisiting some of the previous experiments in a facility with separated upstream water
supply.
The experimental facility consists in a 10 m long, 0.4 m wide and 0.1 m high main
channel in the centre of two symmetrical 0.7 wide floodplains. The transition between the
subsections is made by banks with 45º slope. The slope of the flume bottom is 0.0011 m/m.
The original bottom is hydraulically smooth boundary made of polished concrete. Half of the
experiments were done with the floodplains covered by artificial grass (rough boundary).
Besides other measurement instruments, a Vectrino ADV allowed the measurement of
streamwise and spanwise velocity components, turbulence intensities and Reynolds shear
stresses.
In order to avoid the mass transfer in the beginning of the channel, the upstream water
supply is separated between main channel and floodplains taking into account
recommendations presented in recent literature.
Four different flow conditions were tested, corresponding to uniform flows for relative
depths (ratio of the water depths in the floodplain and in the main channel) approximately
equal to 0.15 and 0.3, for smooth and rough floodplains.
The influence of the relative depth and the floodplain roughness is evaluated and some of
the flow characteristics are presented. It includes the lateral distributions of streamwise
velocity and Reynolds stresses.
Finally, the accuracy of the total cross-section discharge obtained by several 1D methods
is assessed. These methods were the Divided Channel Method, the Coherence Method, the
Integrating Divided Channel Method, the Weighting Divided Channel Method and the
Exchange Discharge Method.