Alternative Method for Determining Manning's Roughness Coefficient Using Two-Point Velocity in Equilibrium and Nonequilibrium Sediment Transport

Understanding flow resistance equations, such as Manning�s roughness equation, is essential for river design and improvement. Estimating Manning�s roughness coefficient becomes more complicated when sediment transport is involved. This study takes an alternative approach by using velocity profiles to examine how sediment transport affects Manning�s roughness coefficient. To achieve this goal, 1200 velocity profiles with sediment-feeding (SF) and nonsediment-feeding (NSF) flows are evaluated to determine the (composite) Manning�s roughness coefficient. Sedimentfeeding flows describe sediment flow under equilibrium conditions, whereas non-sediment-feeding flows represent sediment flow under nonequilibrium conditions. A Sontek 16-MHz Acoustic Doppler Velocimeter is used to measure the velocity (and turbulence) profiles. In addition to the present data, 225 secondary velocity profile data sets are analyzed in this study. The research findings indicate that the composite Manning�s roughness coefficient nco can be determined from Manning�s roughness coefficient nz/B at z/B in the transversal direction, using two points of the velocity profile at y/H = 0.2 and 0.4 in the vertical direction. The differences in the velocity profile shape (u/U) due to sediment feeding, particularly in inner regions (y/H � 0.2), affect the value of nz/B. nco for sediment-feeding flows are generally higher than the crosssection Manning roughness coefficient n. As nco (based on nz/B) is based on the velocity profile, the nco values change with sediment transport. Meanwhile, the n values remain unchanged because the equation variables cannot detect the presence of sediment transport. For non-sediment-feeding flow, the differences in nco with n are 14.80 for a fixed bed (FB) and 18.17 for a movable bed (MB). The differences are even more pronounced for sediment-feeding flow at 33.01 for a fixed bed and 36.52 for a movable bed. The point where nz/B/nco = 1 occurs at z/B = 0.2 from the channel sidewall. This suggests that nz/B, measured at z/B = 0.2 from the channel sidewall, provides a good representation of nco for the section.