Phase diagrams of bed states in steady, unsteady, oscillatory and mixed flows

Abstract

In many flow or wave dominated environments the bed material is moulded in morphological features. In a hydraulic sense, basically two genetic types of forms are distinghuised. First, those generated by interaction of general, changing flow or wave conditions with the bed, such as scroll bars in rivers and nearshore bars in coastal environments. Second, bedforms generated by interaction of the bed with the turbulent or orbital structures of the flow, such as current ripples, dunes and orbital ripples are formed. These bedforms may be predictable from data of appropriate hydraulic and sediment conditions. This paper is about the bedforms. In the past decades a goal of sedimentological studies of bedforms and other bed phases was to develop empirical phase diagrams that show the regimes in which various bed states, e.g., bedforms and sheet flow, are stable. Such diagrams (e.g., Southard and Boguchwal, 1990) are used for two purposes: for assessing the likely bed states in roughly known flow conditions and sediment composition, and to reconstruct the flow conditions from known bed state or sedimentary structures in palaeo- and extraterrestrial environments. In addition, the diagrams direct further research on bed state stability. Phase diagrams of bed states provide a high-altitude view of current ripples, dunes, orbital ripples and other bed states and the conditions in which these occur. They are not intended to replace more detailed models (e.g. Nielsen, 1981, Van Rijn, 1993) which predict not only their emergence and disappearance but also their length or height. Both the phase diagrams and models are empirically well verified, but when plotted together in the bed phase diagrams, there is a striking disagreement. Moreover, bed states in combined waves and currents are much less well documented and understood than in currents only. The aim of this paper is 1) to compare existing phase diagrams with existing bedform and bed state predictors, and 2) to rationalise the diagrams for currents, waves and combined flow based on more physically-based models for bed phases. First, the nondimensional variables along the axes of the diagrams are compared. Next, a short review is given of existing phase diagrams and bed state predictors. A set of new diagrams is then presented for currents, waves and combined flow. Rather than to fill these diagrams with many datasets, they are compared to the lines of existing, empirically well verified diagrams. Finally, the limitations and problems of the new diagrams are discussed, such as the predictability of large bedforms with poor time-adaptation in large water depths and under transient conditions (e.g. floods, tides, storms)