Dune dynamics in a tidal inlet channel of the Danish Wadden Sea

Abstract

Although flow transverse bedforms have been studied for more than a century from field studies, laboratory flume experiments and mathematical models, their initiation, development and dynamics are far from being fully understood.The objective of this study is to investigate the dynamics of dunes on different spatial and temporal scales. The study area is in the Grådyb tidal inlet channel, which forms the navigation channel to Esbjerg, in the Danish Wadden Sea. State of the art swath bathymetry at centimetre-scale resolution and precision is applied in order to observe and quantify processes on both long-term (annual) and short-term (single semi-diurnal tidal cycle) scales. In addition, bed material samples and high-resolution measurements of the flow are obtained in order to relate the dune dynamics to the sediment dynamics and prevailing hydrodynamics. The main results from the investigations are summarised in five papers.The precision of the applied swath bathymetry system under normal field survey conditions is determined by repetitive bathymetric measurements of a shipwreck over three years, showing a precision in all three directions on centimetre-scale (Paper I).Large barchanoid-shaped dunes are shown to develop due to an increase in dune migration from the centre towards the sides of the channel, resulting from a decrease in dune height from the centre towards the sides of the channel. The decrease in dune height is ascribed to an equivalent decrease in grain size from the centre towards the sides of the channel, as water depth and flow velocity are uniform across the channel. Measured annual bedload transport rates are significantly over-predicted by common and widely applied bedload transport formulae (Paper II).Quantifying sediment transport during a single semi-diurnal tidal cycle shows that bedload transport rates are higher on the crests than in the troughs of the large compound dunes. Net dune migration can be flood directed during a single tidal cycle, despite an annual net ebb-directed dune migration. Finally, measured bedload transport rates during single tidal cycles are also significantly over-predicted by common and widely applied bedload transport formulae (Paper III).Relating dune dynamics to hydrodynamics during a single semi-diurnal tidal cycle (same as in Paper III) shows that the higher dune crest than trough mobility is due to higher flow velocities at the dune crest. In addition, bed material goes into suspension with accelerating ebb flow and settles with decelerating ebb flow, resulting in an average erosion and accretion of the bed of ~6.5 cm. During flood tide the bed is practically stable. This results in a potential net export of sediment to the ebb tidal delta during every single semi-diurnal tidal cycle. Furthermore, the height of the compound dunes follows water depth, which acts as a limiting factor to dune growth, whereas the height of the superimposed stoss side dunes follows flow velocity, i.e. superimposed stoss side dunes are water depth-independent (Paper IV).General statements relating dune dimensions directly to the form-corrected Shields parameter are established from the depth-independent superimposed stoss side dunes, showing that dune dimensions are primarily controlled by flow strength and grain size, and that water depth can act as an additional limiting factor to dune growth. This can be used e.g. to give a first approximation of flow conditions, as long as information on grain size and dune dimensions are available, as it is often the case in oceanographic surveys (Paper V).The introduction of high-resolution, high-precision swath bathymetry has truly revolutionised the study of bedform dynamics. The spatial depiction and the possibility to precisely quantify dimensions and dynamics have turned field studies into quasi laboratory experiments. Although it is impossible to steer the controlling parameters (flow velocity and grain size) and limiting factors (water depth), as in flume studies, it is possible to precisely measure flow velocity, grain size and water depth as well as the related bedform dynamics. However, considerable limitations are still present in studies of bedform dynamics, e.g. the absolute quantification of suspended sediment in the water column as well as the determination of near-bed flow velocities without disturbing either the flow or the bed.