Effect of flow discharge and median grain size on mean flow velocity under overland flow

Precise estimation of mean flow velocity (U-mean) is imperative for accurate prediction of hydrographs and sediment yield. For overland flow, U-mean is normally estimated by multiplying the dye or salt based velocity measurement with a correction factor (alpha). A wide range of correction factors is available in the literature, all of which were derived under different experimental conditions. The selection of a suitable a has become a main challenge for accurate mean flow calculations. This study aimed to assess the variability of a with grain size (D-50) and slope (S), and to evaluate the dependency of U-mean on flow rate (Q), D-50 and S by regression analysis. Flume experiments were performed at Q varying from 33 to 1033 x 10(-6) m(3) s(-1), S ranging from 3 degrees to 10 degrees, and D-50 ranging from 0.233 to 1.022 mm. Flow velocities were measured directly with the dye tracing technique (U-dye), and derived indirectly from flow depth measurements (U-depth). The Udepth measurements were considered as U-mean. The derived alpha (U-depth/U-dye) values did not remain constant with sediment size and increase significantly with the increase of D-50. The mean alpha values for 0.230, 0.536, 0.719 and 1.022 mm sands were 0.44, 0.77, 0.82 and 0.82, respectively. Hence, due to the substantial variation of alpha with D-50, no absolute alpha value is applicable to all hydraulic and sedimentary conditions. However, mean alpha values for 0.230, 0.536 and 0.719 mm sands were found comparable with alpha values available in the literature for similar grain sizes. The influence of Q S. and D-50 on U-mean was studied by regression analysis. Regression analysis depicted the significant influence of Q and D-50 on U-mean. while the effect of slope was found to be non-significant. Comparison of the derived regression equation with five literature datasets showed that the model can predict mean flow velocities in overland flow at a reasonable accuracy as long as the mean velocity is below 0.4 m s(-1). At higher velocities the error becomes unacceptably large. (C) 2012 Elsevier B.V. All rights reserved.