Predictive Capability of Bedload Equations Using Flume Data

The study on bedload transport behaviour is widely explored from the last few decades and many semiempirical or empirical equilibrium transport equations are developed. The phenomenon is a very complex due to its varied physical properties like velocity, depth, slope, particle size in the alluvial system. In practical applications, these formulae have appreciable deviation from each other in derivation and also their ranges of applications are different. Here, bedload transports have been categorized into moderate bedload transport and intense bedload transport depending upon the Einstein bedload transport parameter. Based on large database of different bedload measurements, a comparative analysis has been performed to ascertain prediction ability of different bedload equations based on various statistical criteria such as the coefficient of determination, Nash-Sutcliffe coefficient and index of agreement. It has been found that equations based on shear stress have worked better than other approaches (discharge, probabilistic and regression) for flume observations.