Seasonal Variation of Velocity Fields in Lined Channels: Impact on Flow Measurement

The velocity distribution in open-channels is still the core of many investigations. In particular, vegetation influence is being widely studied as it largely affects the flow properties, and it must be taken into accounts in many practical applications such as discharge estimation, flow level prediction, pollutant dispersion, sediment deposition dynamics... Most of these investigation are conducted in laboratory flumes where accurate experimental techniques can be used, such as micro-ADV and PIV. They lead to a better understanding of turbulence-induced forces by vegetation, effect of plant strength and foliage, Monami waves... In natural stream, the effect of vegetation has mainly focused on the influence of vegetation on resistance (Green et al., 2006). Artificial water conveyance systems may be considered as intermediate between laboratory flumes and natural streams since they have generally a regular shape (such as laboratory flumes) but also present a part of the complexity of natural stream, with the presence of natural and non-uniform vegetation. This vegetation may be composed of macrophytes, but smaller-size colonies present in open-channels may also affect velocity profiles, as observed in laboratory flumes (Labiod et al., 2007; Graba et al., 2010). Field observations showed that the development of such vegetation induced very large variations of roughness coefficients within short periods (a few weeks), causing serious management problems (Lozano et al., 2012) and large deviations of rating curves. Measurement techniques based on velocity probes and standard velocity distributions may also be affected by these deviations. The main focus of this paper is the evolution of velocity profiles in real-scale artificial open-channels faced with vegetation growth. The objective is to link the parameters of the velocity distribution function with observed physical characteristics: channel characteristics, vegetation height. We would like to show that the laboratory knowledge on submerged vegetation can help to analyze velocity profiles and to compute discharge for real channel with some simplifications. The analysis is based on field measurements in an irrigation canal performed by micro-ADV during 8 months and on the knowledge from laboratory flow above vegetation. The validity of the modified log-wake law (Guo and Julien, 2008) derived from laboratory flume experiments is studied then. The effect of vegetation growth on roughness, and then velocity distribution, is finally discussed.