River dunes are dynamic and periodic bedforms in sand and gravel bed rivers, which propagate in a streamwise direction. Understanding of river dune dynamics is essential for river management. Dune shape, in combination with celerity, is a proxy for sediment transport (McElroy & Mohrig, 2009;Zomer et al., 2021), which is an important quantity to calibrate morphological models. Such models can support dredging operations to enable safe navigation. Little is known about dune dynamics during low flows due to a focus on dune dynamics during high flows (Julien & Klaassen, 1995;Wilbers, 2003) for roughness and water level predictions (Naqshband et al., 2017; van Duin et al., 2017). Low flow is defined by a sediment transport capacity just above the threshold of motion.River dune dynamics have been studied by a wide range of methods: flume experiments (

No abstract

<p>During low flows, river dunes determine the navigable depth of rivers, influencing the maximum draft of ships. Accurate predictions of the height and location of river dunes during low flows is needed to plan shipping operations in rivers. Currently, little data of dune evolution during low flows is available and analyzed, as most research is focused on high flows and most data is retrieved in flume experiments. However, the scaling from flume to full scale river gives issues with the lee slope angle and secondary bed forms.[LL(1]  Therefore, research on dune evolution on full scale rivers is lacking.</p><p>For this research multibeam echo sounding (MBES) measurements of the Waal river, Netherlands, were used. These measurements were done once per two weeks and cover the fairway. The data was made available by Rijkswaterstaat (Department of public works and water management, Netherlands). We developed a method to analyze dune shape in a large dataset of bed data. This method was applied on a stretch of 2 km Waal river, between the cities of Tiel and St. Andries. The research period covers the whole year 2018. This year is characterized by three separate discharge regimes. High water during January until March, median discharge from April until June and extreme low discharges from July until November.</p><p>In the first step of the data analysis locations of the primary dunes were determined using a wavelet analysis. At these locations, the dune crests and troughs were identified. With the crests and troughs, the shape characteristics such as dune length, height lee slope angle and propagation speed were determined. The dune characteristics were eventually related to the governing discharge.</p><p>The first results show that the river dunes are mobile during extreme low flows. After a transition period of one month, where the discharge drops from the median value towards the low discharge, the dune length and height become statistically stable. While the dune shape in flow direction becomes stable during these low flows the three dimensionality increases, not only in primary dune shape but also the appearance of secondary dunes and ripples. The dune height near the right bank is smaller than in the middle of the river, towards the left bank the height decreases again. The differences between the banks and the middle of the fairway increase as discharge decreases. Also, by visually inspecting the bed profiles at other locations, a similar trend is observed.</p><p>The first results show that the location in the river cross-section influences the dune characteristics. These differences increase as the discharge decreases. In further work we will extend the research area over the full length of the Waal river, to give a quantitative analysis supporting our visual results and to include influence of sediment size. As the differences in the cross-section can be found throughout the river, we will also investigate the influence of shipping on the differences in dune shape.</p>

<p>River dune modelling ranges from linear stability analysis to analyse the initial growth of the dunes (Fredsøe, 1983) up to three dimensional numerical models which can simulate the dune evolution by modelling the sediment transport on particle level (Nabi et al., 2013). For engineering purposes, such as efficient planning of dredging operation or dynamic modelling of dune roughness for water level predictions, a quick and accurate dune development model is needed. Therefore we further develop the model of Paarlberg et al. (2009), in order to accurately model dune shape and migration during high, median and low flow situations.</p><p>This model simulates dune development using a flow module in a two dimensional vertical plane and a bed load transport module which calculates the bulk transport. The model solves the flow over the domain of one dune length, using cyclic boundary conditions. The domain length, covering one dune length, is determined using a numerical linear stability analysis. It has been proven to accurately and fairly quickly reproduce the dune height of flume experiments and it is also able to simulate the transition to upper stage plane bed accurately (Duin et al., 2021).</p><p>However, for low flow situations it has not been validated yet. One of the main issues during low flow is that the relation between water depth and dune length is not linear and the adaptation of the dune length to new, smaller, water depths and flow velocities is not instantaneous (Lokin et al., 2022). The linear stability routine determines the dune length to which the dunes will grow based on a plane bed with a small disturbance, and directly updates the domain length to this newly determined dune length. In this research we have investigated options to incorporate the lag in the dune length adjustment during the falling stage of a flood wave. Implementing a lag in the dune length adjustment, such that the dune length adapts at a rate that is linked to the depth averaged flow velocity, leads to more realistic dune lengths.</p><p>Duin, O. J. M. van, Hulscher, S. J. M. H., & Ribberink, J. S. (2021). Modelling Regime Changes of Dunes to Upper-Stage Plane Bed in Flumes and in Rivers. <em>Applied Sciences 2021, Vol. 11, Page 11212</em>, <em>11</em>(23), 11212. https://doi.org/10.3390/APP112311212</p><p>Fredsøe, J. (1983). Shape and dimensions of ripples and dunes. <em>Mechanics of Sediment Transport. Proc. Euromech 156, Istanbul, July 1982</em>.</p><p>Lokin, L. R., Warmink, J. J., Bomers, A., & Hulscher, S. J. M. H. (2022). <em>River dune dynamics during low flows</em>. https://doi.org/submitted for publication</p><p>Nabi, M., De Vriend, H. J., Mosselman, E., Sloff, C. J., & Shimizu, Y. (2013). Detailed simulation of morphodynamics: 3. Ripples and dunes. <em>Water Resources Research</em>, <em>49</em>(9), 5930–5943. https://doi.org/10.1002/wrcr.20457</p><p>Paarlberg, A. J., Dohmen-Janssen, C. M., Hulscher, S. J. M. H., & Termes, P. (2009). Modeling river dune evolution using a parameterization of flow separation. <em>Journal of Geophysical Research: Earth Surface</em>, <em>114</em>(1). https://doi.org/10.1029/2007JF000910</p>