Resolving river planform and width effects on flow in meandering rivers for optimal placing of ADCP cross sections

Acoustic instruments such as multibeam echosounders (MBES) and acoustic Doppler current profilers (ADCPs) can collect flow and bathymetric data at high spatial and temporal resolutions. Researchers are widely using MBES and ADCP instruments to advance our understanding of meandering river morphodynamics. Whereas MBES instrumentation produces a continuous bathymetric surface, ADCP usage acquires the three-dimensional flow field at a series of cross sections. The spacing of ADCP cross sections has generally been based on a rule of thumb (for example, a multiple of the channel width), but is constrained by the available survey time. Theoretical advances regarding the morphologic adjustment of meandering rivers over the past 50 years have identified two important parameters: channel width-to-depth ratio (W/H) and channel radius of curvature-to-width ratio (R/W). Though, width is a controlling factor, it is its ratios with respect to channel depth and radius of curvature, that are important. In addition, channel curvature and width may vary considerably along the river. Hence, determining ADCP cross section locations based on a multiplication factor of channel width is challenging. We used a suite of wavelet analysis methods to determine the effects of channel curvature and width on flow measured using ADCPs along the Pearl River, LA/MS, Little Tallahatchie River, MS, and Wabash River, IL/IN. For example, for six bends on the Pearl River, curvature varied between 0.01 and -0.01 and top width varied between 80 and 180 m. The analysis showed that the streamwise velocity along the channel centerline was controlled by variations in channel width for most of the study reach. The primary extremes in near-bank streamwise velocity were controlled by channel curvature, but secondary peaks at some compound bends were controlled by changes in channel top width rather than local increases in curvature.  These findings are used to develop a tool combining a constant-width linear model of meandering-river flow with streamwise channel width profiles to determine the optimal placing of ADCP cross sections. Such information is specifically useful for rivers, such as the Little Tallahatchie River, that have a small width-to-depth ratio and a large channel radius of curvature-to-width ratio. To collect ADCP data using spacing equal to channel width for such rivers would lead to unrealistic survey duration.