1 INTRODUCTION Field measurements of river velocity can provide a valuable contribution to understanding morpho- logical processes, contaminant transport, and stream ecology. The acoustic Doppler current pro- filer (ADCP) is used increasingly in river-related studies to measure velocity and determine flow rate (Simpson 2001), turbulence characteristics (e.g. Stacey et al. 1999, Nystrom et al. 2007), boundary shear stress (Sime et al. 2007), and se- diment transport (Rennie & Millar 2004). Addi- tionally, the high spatial resolution data from ADCPs may provide a useful tool for calibrating and validating computational fluid dynamics models. This study presents ADCP measurements from the lower Roanoke River, a regulated river in eastern North Carolina, USA and describes a procedure for determining local boundary shear stress. Fixed-vessel measurements (Muste et al. 2004) were obtained at a location within a meand- er bend for two flow rates, one close to the mean annual flow (flow rate, Q = 220 m 3 s -1 ) and the other at near bankfull conditions (Q = 565 m 3 s -1 ). Maintaining a fixed location within the river for the entire measurement duration presented a chal- lenge. The effect of the ADCP motion on the measured velocity profiles is assessed. Velocity profiles are often used as an indirect method to determine mean boundary shear stress in natural rivers (Wilcock 1996). While several methods are available to determine the time- averaged local boundary shear stress (e.g. Biron et al. 2004, Dietrich & Whiting 1989), this study employs the theoretical log-law, given as: ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ = o z z u u ln 1 * κ (1) where u = velocity, * u = shear velocity ( * u = (τ o /ρ) 0.5 , where τ o = boundary shear stress and ρ = fluid density), κ = von Karman’s constant (κ = 0.40), z = position perpendicular to the channel bed, and z o = roughness height. Following the ap- proach of Raupach et al. (1991), the perpendicular position above the bed is defined as z = Z + d, where Z = position above the origin as defined by the top of the roughness elements and d = zero displacement plane. When a measured velocity profile is available, a least squares error approach can be used to fit a linear equation to the profile of u vs. ln(z). This approach has two primary advan- tages: (i) no knowledge of the roughness height is required to determine the shear velocity and (ii) a measure of the goodness of fit for the data is available through the coefficient of determination, also known as the R 2 -value. The least square er- Local boundary shear stress estimates from velocity profiles measured with an ADCP J. Petrie, P. Diplas & S. Nam Baker Environmental Hydraulics Laboratory, Department of Civil & Environmental Engineering, Virginia Tech, USA M. S. Gutierrez Division of Engineering, Colorado School of Mines, USA ABSTRACT: The acoustic Doppler current profiler (ADCP) has become an important tool in the study of river processes. When measurements are obtained at a fixed location within the river channel, time- averaged velocity profiles can be calculated. These profiles have the potential to quantify flow properties such as secondary currents and boundary shear stress. Velocity profiles from ADCP measurements ob- tained on the lower Roanoke River in the USA are used to estimate local mean boundary shear stress. The procedure combines the well known log-law with visually establishing the region within the flow depth where this law is valid. Additionally, methods are presented to (i) determine if movement of the ADCP adversely affects the measured velocity profile, (ii) test whether the recorded data is stationary, and (iii) calculate the depth-averaged velocity. Keywords: Acoustic techniques, Boundary shear, Field tests, Stationary processes, Velocity profile River Flow 2010 - Dittrich, Koll, Aberle & Geisenhainer (eds) - © 2010 Bundesanstalt für Wasserbau ISBN 978-3-939230-00-7 1749