No abstract

Current methods employed by the United States Geological Survey (USGS) to measure river discharge are manpower intensive, expensive, and during high flow events require field personnel to work in dangerous conditions. Indirect methods of estimating river discharge, which involve the use of extrapolated rating curves, can result in gross error during high flow conditions due to extrapolation error and/or bathymetric change. Our goal is to develop a remote method of monitoring volumetric discharge that reduces costs at the same or improved accuracy compared with current methods, while minimizing risk to field technicians. We report the results of Large-Scale Particle Image Velocimetry (LSPIV) and Acoustic Doppler Velocimetry (ADV) measurements conducted in a wide-open channel under a range of flow conditions, i.e., channel aspect ratio (B/H 5 6.6-31.9), Reynolds number (Re H 5 4,950-73,800), and Froude number (Fr 5 0.04-0.46). Experiments were carried out for two different channel cross sections (rectangular and asymmetric compound) and two bathymetric roughness conditions (smooth glass and rough gravel bed). The results show that the mean surface velocity normalized by the depth-averaged velocity (the velocity index) decreases with increasing d*/H, where d* is the boundary layer displacement thickness and that the integral length scales, L 11,1 and L 22,1 , calculated on the free-surface vary predictably with the local flow depth. Remote determination of local depth-averaged velocity and flow depth over a channel cross section yields an estimate of volumetric discharge.

Synoptic information on bed shear stress is necessary in predicting the transport of sediments and environmental contaminants in rivers and open channels. Existing methods of estimating bed shear stress typically involve measuring vertical profiles of streamwise velocity or Reynolds stress and taking advantage of the logarithmic or the constant stress region, respectively, to determine friction velocity and subsequently, bed shear stress. While effective, these methods yield local measurements of bed shear stress only. Direct measurements of bed shear stress can also be obtained through measurements with a drag plate. However, this method yields average shear stress information over the area of the plate and is impractical for large‐scale implementation in the field. Here we present a method capable of providing continuous synoptic measurements of bed shear stress over a large field‐of‐view. A series of Large‐Scale Particle Image Velocimetry (LSPIV) and Acoustic Doppler Velocimetry (ADV) measurements were made in a variety of flows generated in a wide‐open channel facility. Turbulent dissipation is calculated on the free surface from the 2‐D LSPIV results and is correlated with near‐surface ADV measurements of turbulent dissipation in the water column. The ADV results are consistent with the Nezu (1977) established relationship for the vertical variation of turbulent dissipation in the water column. Knowledge of the correlation between free‐surface and near‐surface dissipation values coupled with Nezu's (1977) relationship allow a robust and accurate estimate of friction velocity to be made and subsequently, shear stress at the bed can be estimated.

No abstract