Predicting boundary shear stress and sediment transport over bed forms

To estimate bed-load sediment transport rates in flows over bed forms such as ripples and dunes, spatially averaged velocity profiles are frequently u sed to predict mean boundary shear stress. However, such averaging obscures the complex, nonlinear interaction of wake decay, boundary-layer developme nt, and topographically induced acceleration downstream of flow separation and often leads to inaccurate estimates of boundary stress, particularly sk in friction, which is critically important in predicting bed-load transport rates. This paper presents an alternative methodology for predicting skin friction over 2D bed forms. The approach is based on combining the equation s describing the mechanics of the internal boundary layer with semiempirica l structure functions to predict the velocity at the crest of a bedform, wh ere the flow is most similar to a uniform boundary layer. Significantly, th e methodology is directed toward making specific predictions only at the be d-form crest, and as a result it avoids the difficulty and questionable val idity of spatial averaging. The model provides an accurate estimate of the skin friction at the crest where transport rates are highest. Simple geomet ric constraints can be used to derive the mean transport rates as long as b ed load is dominant.