Sediment transport phenomenon in rivers: an alternative perspective

Abstract During the last century, a great deal of research has been devoted to modeling the dynamics of sediment transport phenomenon. In spite of the progress achieved, our understanding of the sediment transport phenomenon is far from complete, as there is no generally accepted relationship between the components involved, e.g. water discharge, suspended sediment concentration and bed load. Also, any error (e.g. measurement error) in one component of the system (e.g. discharge) could eventually lead to an inaccurate outcome in deriving either another component or relationships between the components. One possible way to avoid these problems is by modeling the component of interest (e.g. bed load) using a time series of the same component itself. Such an approach may also be able to represent the dynamics of the entire system. In this regard, the concept of phase-space reconstruction, i.e. reconstruction of a single-variable series in a multi-dimensional phase-space, and the related ideas of chaos theory could be useful. This study investigates the possible use of such an approach for understanding bed load dynamics. The approach is employed for predicting the bed load dynamics in the Mississippi River basin in USA. The predictions are made using a local approximation method. The results indicate very good agreement between the predicted and the observed bed load values. The near-accurate predictions indicate the appropriateness of phase-space reconstruction and local approximation prediction for understanding the bed load dynamics. The results also reveal that the bed load dynamics are dominantly influenced by three variables, suggesting that the dynamics could be understood from a low-dimensional chaotic dynamical perspective.