Sedimentology and morphology of a low-sinuosity river: Calamus River, Nebraska Sand Hills

A study reach of the Calamus River, Nebraska Sand Hills, has a low sinuosity (less than 1.3) and braiding parameter (less than 1). Depending on sinuosity, the channel is occupied by alternate bars and point bars, the emergent parts of which form nuclei for midstream bars (islands). Channel migration occurs by bend expansion and translation, downstream and lateral growth of islands, and by chute cutoff. Channel-bed sediment is mainly medium-grained sand, but gravel and coarser sand sizes occur in thalweg areas adjacent to cutbanks and upstream parts of bars and islands, and finer sands occur on the downstream parts of bars and filling channels. Curved-crested dunes cover most of the channel bed at most flow stages, with ripples restricted to shallow areas near banks. Bed material is mostly large-scale cross-stratified, with small-scale cross-strata interbedded with plant debris occurring in topographically high areas near banks. Vibracores through channel bars show a basal erosion surface overlain by large-scale cross-stratified sands, in turn overlain by small-scale cross-stratified sand interbedded with plant debris. The overall sequence generally fines upwards, but the large-scale cross-stratified portion either fines upwards, coarsens upwards, or shows little grain size variation. Lithofacies distributions vary spatially within and between bars depending on position in the bar and local channel curvature/width, in a similar way to unbraided rivers elsewhere. Lithofacies of bar deposits are similar to those in the active channel, and the elevations of the basal erosion surface and adjacent channel thalweg correspond closely. Channels abandoned by chute cutoff are filled progressively from the upstream end, and comprise deposits similar to the downstream parts of bars (i.e. fining upwards). The downstream extremities of channel fills may contain large proportions of peat relative to sand, but little mud due to the paucity of such fine suspended load in the Calamus.     

An empirical model for the development and equilibrium morphology of current ripples in fine sand

A flume study on the development and equilibrium morphology of current ripples in fine sand (D₅₀ = 0·238 mm) was performed to extend an empirical model for current ripple stability in 0·095 mm sand to larger grain sizes. The results of the flume experiments agree with the very fine sand model that current ripple development from a flat bed is largely independent of flow velocity. At all flow velocities, ripples evolve from incipient, through straight, sinuous and non-equilibrium linguoid, to equilibrium linguoid plan morphology. The time needed to achieve an equilibrium linguoid plan form is related to an inverse power of flow velocity and ranges from several minutes to more than hundreds of hours. Average equilibrium height and length are 17·0 mm and 141·1 mm respectively. These values are about 20% larger than in very fine sand. Equilibrium ripple height and length are proportional to flow velocity near the stability field of dunes. In the same velocity range, a characteristic grouping of ripples with smaller ripples migrating on the upstream face of larger ripples was observed. Bed-form development shows a conspicuous two-phase behaviour at flow velocities < 0·49 m s^?1. In the first phase of development, ripple height and length increase along an exponential path, similar to that at higher flow velocities, thus reaching intermediate equilibrium values of 14·8 mm and 124·5 mm respectively. After some time, however, a second phase commences, that involves a rapid increase in bed-form size to the typical equilibrium values for 0·238 mm sand. A comparison with literature data shows that the results obtained for 0·238 mm sand agree reasonably well with other flume studies at similar grain size. Yet considerable variability in the relationships between ripple dimensions and flow strength ensues from, among others, underestimation of equilibrium time, shallow flow depths and differences in sediment texture.     

Response of sand ripples to change in oscillatory flow

Ripples take time to evolve to a new equilibrium state in response to a change in wave-generated oscillatory flow. The paper presents results from flow tunnel experiments designed to examine oscillatory flow transient ripple processes under controlled, full-scale laboratory conditions. The experiments include study of the growth of ripples from flat bed and the evolution of existing ripples to new equilibrium ripples in response to a step change in the flow. In general, ripples evolve through a combination of two main processes: (i) from a flat bed or from a bed consisting of ripples that are smaller than the equilibrium ripples through a combination of 'slide' and 'merge'; (ii) from a bed consisting of ripples that are larger than the equilibrium ripples through a combination of 'split' and 'merge'. The experimental results show that equilibrium ripple geometry is independent of initial bed morphology while the time to reach equilibrium is largely independent of the initial bed and the equilibrium ripple size. The time to reach equilibrium depends strongly on the mobility number, and a new empirical equation relating mobility number and the number of flow cycles to equilibrium is proposed. This equation is combined with a simple exponential function for ripple height growth or decay to produce a new empirical model for ripple height evolution, which gives a reasonably good overall agreement with the measurements. The model is based on experiments involving one sediment size only and further work is needed to develop the model for other sand sizes.     

浅水非线性波作用下沙纹床面底层流动特性试验研究

通过水槽试验研究浅水非线性波作用下沙纹床面底层流动特性,利用CCD图像技术观测分析非对称沙纹的形成和演化规律。利用声学多普勒测速仪(ADV)测量非对称沙纹底床上的流场,得到了不同波高、周期、水深条件下的沙纹峰顶和谷底断面的瞬时速度。试验结果分析表明,浅水非线性波作用下床面上形成非对称沙纹,其近底流速具有较强紊动特性,随着距床面距离的增大紊动强度逐渐减弱。在水流方向改变时,沙纹背部具有明显漩涡运动。沙纹背后形成的漩涡能起到维持沙纹的作用。浅水非线性波作用下,沙纹的形成原因主要是床面泥沙颗粒在非对称流动和床面近壁粘性底层中漩涡结构动力作用下,作受迫摆动、推移所致。     

Time-sequence observations of wave-formed sand ripples on an ocean shoreface

Analysis of an 18-day time-lapse film record of shoreface ripple development, with concurrent measurements of near-bottom flow and surface waves, provides new insight on equilibrium bedform conditions, adjustment of ripple planform to variable hydrodynamics, and ripple migration behaviour. The study was conducted in approximately 10 m water depth, 1 km off Martinique Beach on the Atlantic coast of Nova Scotia (Canada), under low-energy summer wave conditions. Significant wave-height and peak period during the study averaged 0–7 m and 8 s, respectively, with extremes up to 1–7 m and 11 s during passage of three weak weather disturbances. Six mutually exclusive ripple types have been defined: (1) short-wavelength regular ripples; (2) variable bifurcated ripples; (3) variable terminated ripples; (4) short-crested ripples; (5) long-wavelength regular ripples; and (6) chaotic ripples. Ripple wavelength ranged from 0–07 m to 0–24 m and displayed a strong Reynolds number dependence. Together with other published field data, the results suggest a lower limit of γ=0–06 m for the wavelength of wave ripples in ocean shoreface environments. Ripple orientation ranged through 38° and responded rapidly to changes in wave approach direction, but did not conform to the orientation of the adjacent shoreline. Ripples were observed to migrate both on- and off-shore (with and against the wave advance direction) at rates up to ±0–1 m h^-1, associated with net flows other than wave-induced onshore asymmetry and mass transport. Migration (mainly of ripple types 1 and 2) occurred during the peak of storm events, but showed no obvious correlation with measured near-bottom flow magnitude or direction. Ripple behaviour demonstrates equilibrium with prevailing dynamic conditions when straight-crested rippie types 1 and 5 are present. Disequilibrium in orientation or dimensions is expressed by increasing sinuosity, bifurcation and crest termination in types 2,3,4 and 6.     

Flow and turbulence structure across the ripple–dune transition: an experiment under mobile bed conditions

Current knowledge of flow and turbulent processes acting across the sand bed continuum is still unable to unequivocally explain the mechanism(s) by which ripples become dunes. Understanding has been improved by comparative high-resolution studies undertaken over fixed bedforms at different stages in the continuum. However, these studies both ignore the role of mobile sediment and do not examine flow structure during the actual transition from ripples to dunes. The aims of the paper are: (i) to describe flow and turbulence characteristics acting above mobile bedforms at several stages across the transition; and (ii) to compare these data with those arising from experiments over fixed ripples and dunes. Laboratory experiments are presented that examine the turbulence structure across seven distinct stages of the transition from ripples to dunes. Single-point acoustic Doppler velocimeter sampling at three flow heights above a developing mobile boundary was undertaken. Time-averaged statistics and the instantaneous quadrant record reveal distinct changes in flow structure either side of the change from ripples to dunes. Initially, shear-related, high-frequency vortex shedding dominates turbulence production. This increases until two-dimensional (2D) dunes have formed. Thereafter, turbulence intensities and Reynolds stress decline and three-dimensional dunes exhibit values found over 2D ripples. This is the result of shear layer dampening which occurs when the topographically-accelerated downstream velocity increases at a faster rate than flow depth. Activity at reattachment increases due to high velocity fluid imparting high mass and momentum transfer at the bed and/or wake flapping. Suspended sediment may also play a role in turbulence dampening and bed erosion. Ejections dominate over sweeps in terms of event frequency but not magnitude. Strong relationships between inward interactions and sweeps, and ejections and outward interactions, suggest that mass and momentum exchanges are dependent upon activity in all four quadrants. The results contradict the notion present in most physical models that larger bedforms exhibit most shear layer activity. Consequently an improved model for the ripple–dune transition is proposed.     

Depositional processes,bedform development and hybrid bed formation in rapidly decelerated cohesive (mud–sand) sediment flows

Flows with high suspended sediment concentrations are common in many sedimentary environments, and their flow properties may show a transitional behaviour between fully turbulent and quasi‐laminar plug flows. The characteristics of these transitional flows are known to be a function of both clay concentration and type, as well as the applied fluid stress, but so far the interaction of these transitional flows with a loose sediment bed has received little attention. Information on this type of interaction is essential for the recognition and prediction of sedimentary structures formed by cohesive transitional flows in, for example, fluvial, estuarine and deep‐marine deposits. This paper investigates the behaviour of rapidly decelerated to steady flows that contain a mixture of sand, silt and clay, and explores the effect of different clay (kaolin) concentrations on the dynamics of flow over a mobile bed, and the bedforms and stratification produced. Experiments were conducted in a recirculating slurry flume capable of transporting high clay concentrations. Ultrasonic Doppler velocity profiling was used to measure the flow velocity within these concentrated suspension flows. The development of current ripples under decelerated flows of differing kaolin concentration was documented and evolution of their height, wavelength and migration rate quantified. This work confirms past work over smooth, fixed beds which showed that, as clay concentration rises, a distinct sequence of flow types is generated: turbulent flow, turbulence‐enhanced transitional flow, lower transitional plug flow, upper transitional plug flow and a quasi‐laminar plug flow. Each of these flow types produces an initial flat bed upon rapid flow deceleration, followed by reworking of these deposits through the development of current ripples during the subsequent steady flow in turbulent flow, turbulence‐enhanced transitional flow and lower transitional plug flow. The initial flat beds are structureless, but have diagnostic textural properties, caused by differential settling of sand, silt and cohesive mud, which forms characteristic bipartite beds that initially consist of sand overlain by silt or clay. As clay concentration in the formative flow increases, ripples first increase in mean height and wavelength under turbulence‐enhanced transitional flow and lower transitional plug‐flow regimes, which is attributed to the additional turbulence generated under these flows that subsequently causes greater lee side erosion. As clay concentration increases further from a lower transitional plug flow, ripples cease to exist under the upper transitional plug flow and quasi‐laminar plug flow conditions investigated herein. This disappearance of ripples appears due to both turbulence suppression at higher clay concentrations, as well as the increasing shear strength of the bed sediment that becomes more difficult to erode as clay concentration increases. The stratification within the ripples formed after rapid deceleration of the transitional flows reflects the availability of sediment from the bipartite bed. The exact nature of the ripple cross‐stratification in these flows is a direct function of the duration of the formative flow and the texture of the initial flat bed, and ripples do not form in cohesive flows with a Reynolds number smaller than ca 12 000. Examples are given of how the unique properties of the current ripples and plane beds, developing below decelerated transitional flows, could aid in the interpretation of depositional processes in modern and ancient sediments. This interpretation includes a new model for hybrid beds that explains their formation in terms of a combination of vertical grain‐size segregation and longitudinal flow transformation.     

THE NATURE AND ORIGIN OF BED-FORM HIERARCHIES

The bed forms generated by liquid flows over loose-grain boundaries assume only two orientations relative to flow, being either transverse or longitudinal, but range enormously in physical scale. Two types of bed form, current ripples and parting lineations, take a characteristic length (2) which is independent of boundary conditions as expressed by flow depth (d). We have that: λ^L≤ 0.025 m (parting lineations) λ^R≤ 0.60 m (ripples). The characteristic length of other kinds of bed form is correlated with flow depth, according to the equations: λ^D= 1.16D1.^1.55 (dunes) λ^A 6.3d (antidunes, Fr = 1.0) λ^SR= 1.35d^1.31 (sand ribbons) λ^M - 672d^1.11 (meander bars) λ^T= 116d² (tidal current ridges) From these relationships can be deduced the compositions of theoretical hierarchies of bed forms, where a hierarchy is defined as a series of bed forms graded as to size which in combination are adapted to a single overall flow condition. The compositions of these hierarchies, which are closely matched by naturally occurring hierarchies, are broadly related to available flow depth, general flow conditions (lower or upper regime), and availability of sediment. Bed forms are considered to be due to unstable interactions between the bed material and the over-bed flow, of liquid or grains or both. Hierarchies of bed forms arise because the quantities determining the flow are suficiently numerous that several mutually unstable combinations can exist, each combination being expressed in terms of a bed form of a characteristic physical scale and orientation relative to flow.     

Time as an independent variable for current ripples developing towards linguoid equilibrium morphology

Flume experiments show that current ripples on very fine sand surfaces always develop towards a linguoid shape with constant height and wavelength provided that sufficient time is allowed for their formation. Straight and sinuous current ripples only reflect intermediate stages in ripple development and may be regarded as non-equilibrium bedforms. The time period which current ripples require to reach linguoid equilibrium morphology is related to an inverse power of flow velocity. In the transitional stage from current ripples to upper stage plane bed (i.e. washed-out ripple stage) only the equilibrium wavelength remains constant, whereas equilibrium height rapidly decreases to zero. Our observations imply that bed-roughness parameters in sediment transport calculations can be simplified when equilibrium conditions are attained, and that inferences about flow energy from the dimensions of current ripples in very fine sand need to be regarded with caution.     

An experimental study of flow, bedload transport and bed topography under conditions of erosion and deposition and comparison with theoretical models

The nature of flow, sediment transport and bed texture and topography was studied in a laboratory flume using a mixed size-density sediment under equilibrium and non-equilibrium (aggradational, degradational) conditions and compared with theoretical models. During each experiment, water depth, bed and water surface elevation, flow velocity, bed shear stress, bedload transport and bed state were continuously monitored. Equilibrium, uniform flow was established with a discharge of about 0.05 m³ s^?1, a flow depth of about 0.01 m, a flow velocity of about 0.81–0.88 m s^?1, a spatially averaged bed shear stress of about 1.7–2.2 Pa and a sediment transport rate of about 0.005–0.013 kg m^?1 s^?1 (i.e. close to the threshold of sediment transport). Such equilibrium flow conditions were established prior to and at the end of each aggradation or degradation experiment. Pebble clusters, bedload sheets and low-lying bars were ubiquitous in the experiments. Heavy minerals were relatively immobile and occurred locally in high concentrations on the bed surface as lag deposits. Aggradation was induced by (1) increasing the downstream flow depth (flume tilting) and (2) sediment overloading. Tilt-induced aggradation resulted in rapid deposition in the downstream half of the flume of a cross-stratified deposit with downstream dipping pebbles (pseudo-imbricated). and caused a slight decrease in the equilibrium mean water surface slope and total bedload transport rate. These differences between pre- and post-aggradation equilibrium flow conditions are due to a decrease in the local grain roughness of the bed. Sediment overloading produced a downstream fining and thinning wedge of sediment with upstream dipping pebbles (imbricated), whereas the equilibrium flow and sediment transport conditions remained relatively unchanged. Degradation was induced by (1) decreasing the downstream flow depth (flume tilting) and (2) cutting off the sediment feed. Tilt-induced degradation produced rapid downstream erosion and upstream deposition due to flow convergence with little change to the equilibrium flow and sediment transport conditions. The cessation of sediment feed produced degradation and armour development, a reduction in the mean water surface slope and flow velocity, an increase in flow depth, and an exponential decrease in bedload transport rate as erosion proceeded. A bedload transport model predicted total and fractional transport rates extremely well when the coarse-grained (or bedform trough) areas of the bed are used to define the sediment available to be transported. A sediment routing model, MIDAS, also reproduced the equilibrium and non-equilibrium flow conditions, total and fractional bedload transport rates and changes in bed topography and texture very well.     

Scour holes and ripples occur below the hydraulic smooth to rough transition of movable beds

Scour holes often form in shallow flows over sand on the beach and in morphodynamic scale experiments of river reaches, deltas and estuarine landscapes. The scour holes are on average 2 cm deep and 5 cm long, regardless of the flow depth and appear to occur under similar conditions as current ripples: at low boundary Reynolds numbers, in fine sand and under relatively low sediment mobility. In landscape experiments, where the flow is only about 1 cm deep, such scours may be unrealistically large and have unnatural effects on channel formation, bar pattern and stratigraphy. This study tests the hypotheses that both scours and ripples occur in the same conditions and that the roughness added by sediment saltation explains the difference between the ripple–dune transition and the clear‐water hydraulic smooth to rough transition. About 500 experiments are presented with a range of sediment types, sediment mobility and obstructions to provoke scour holes, or removal thereof to assess scour hole persistence. Most experiments confirm that ripples and scour holes both form in the ripple stability field in two different bedform stability diagrams. The experiments also show that scours can be provoked by perturbations even below generalized sediment motion. Moreover, the hydraulic smooth to rough transition modified with saltation roughness depending on sediment mobility was similar in magnitude and in slope to ripple–dune transitions. Given uncertainties in saltation relations, the smooth to rough transitions modified for movable beds are empirically equivalent to the ripple–dune transitions. These results are in agreement with the hypothesis that scours form by turbulence caused by localized flow separation under low boundary Reynolds numbers, and do not form under generalized flow separation over coarser particles and intense sediment saltation. Furthermore, this suggests that ripples are a superposition of two independent forms: periodic bedforms occurring in smooth and rough conditions plus aperiodic scours occurring only in hydraulic smooth conditions.     

Initiation of ripple marks under oscillating water

It was observed that a monolayer of glass beads which were scattered sparsely on a rigid plane floor grew into regular waves of particles under oscillatory water flow. The relative displacement of two nearby particles due to viscous fluid forces seems to be responsible for the initiation of these particle waves. It was also observed that the similar particle waves were formed on the initially flat surface of a thick sand bed and subsequently developed into oscillatory sand ripples of a common type. On the basis of these observations, it is suggested that the particle waves may be the basic cause of the initiation of general ripple marks under oscillatory flow.     

Ripple formation in combined transdirectional steady and oscillatory flow

Measurements are described of the geometry of ripples formed on beds of sand exposed to a steady current at right angles to an oscillatory flow. Four different sands were studied. The oscillation was produced by an oscillating tray set into the bed of a steady-flow flume. It was observed that straight-crested ripples formed by oscillatory flow would usually develop a ‘serpentine’ form when the superimposed steady current exceeded a certain limit. For amplitudes of the tray velocity U_∞ less than about 0.38 m s^-1 this limit corresponded to U_∞/ū_*c>31, where ū_*c is the shear velocity measured just upstream of the oscillating tray. It is suggested that the serpentine form is caused by the interaction of vortices carried back and forth between adjacent ripples. On this assumption, the wavelength of the serpentine form would be proportional to the product of period of oscillation and near-bed steady current velocity. The present measurements appear to support this hypothesis although there is also evidence that the wavelength is influenced by preferred spacing patterns between vortices. The measurements also show the ratio of the amplitude of the serpentine form to its wavelength to be approximately constant. Empirical relationships are derived relating ripple geometry to flow and sediment properties. It is observed that the influence of Reynolds number and sediment properties on the geometry is very weak. It is suggested that this is typical of ripples formed with relatively low sediment transport rates. It is also found that, under the present experimental conditions, the ripple spacing in the direction of oscillation is almost independent of the magnitude of the steady current and in close agreement with the wavelengths previously measured in an oscillating water tunnel. This suggests that the additional inertia effects associated with oscillating tray rigs were not sufficient to affect bed geometry under the present test conditions.     

Zur Geometrie von Wellenrippeln in Sanden unterschiedlicher Korngröße

Field research of wave generated bed forms within complex sediment size distributions near the inlet of a tidal lagoon at the northern coast of Brittany has stimulated an experimental study in a laboratory wave tank. Several sediment mixtures, most of them with bimodal grain size distributions, were exposed to different monochromatic shallow water waves. The observations and measurements included the dynamics of the water waves and the generation, shape, and size of oscillatory bed forms. The experiments confirm the known relationship between grain size and ripple size. In addition it is shown that coarse sand, added to a preexisting fine bed material leads to an increasing asymmetry of ripples. There is some suggestion that the variability of ripple heights is reduced by higher contents of coarse sand. Bimodal sediment size distributions obviously do not cause unusual geometry of ripples — at least within the range of the experimental tests. The different sand size modes move together in one phase, forming structures with more or less homogeniously distributed bed material. Differentiation of sediment sorting does of course occur, but this is in the range of the whole test section. Finally the experiments allowed to test the validity of some wave formulas. The own experiments are compared with some results from field and laboratory studies of other authors.     

Microbial influence on erosion,grain transport and bedform genesis in sandy substrates under unidirectional flow

Interpreting the physical dynamics of ancient environments requires an understanding of how current‐generated sedimentary structures, such as ripples and dunes, are created. Traditional interpretations of these structures are based on experimental flume studies of unconsolidated quartz sand, in which stepwise increases in flow velocity yield a suite of sedimentary structures analogous to those found in the rock record. Yet cyanobacteria, which were excluded from these studies, are pervasive in wet sandy environments and secrete sufficient extracellular polysaccharides to inhibit grain movement and markedly change the conditions under which sedimentary structures form. Here, the results of flume experiments using cyanobacteria‐inoculated quartz sand are reported which demonstrate that microbes strongly influence the behaviour of unconsolidated sand. In medium sand, thin (ca 0·1 to 0·5 mm thick) microbial communities growing at the sediment–water interface can nearly double the flow velocity required to produce the traditional sequence of ripple→dune→plane‐bed lamination bedforms. In some cases, these thin film‐like microbial communities can inhibit the growth of ripples or dunes entirely, and instead bed shear stresses result in flip‐over and rip‐up structures. Thicker (ca≥1 mm thick) microbial mats mediate terracing of erosional edges; they also, foster transport of multi‐grain aggregates and yield a bedform progression consisting of flip‐overs→roll‐ups→rip‐ups of bound sand.     

The development of small scale bedforms in tidal environments: an empirical model for unsteady flow and its applications

Dimensions and plan morphology of current ripples are generally considered to vary with flow velocity and grain size. Recently, however, it has been shown that for sand of D₅₀=0.095 and 0.238 mm the equilibrium dimensions are identical at all velocities within the stability field of ripples and that the plan form of equilibrium ripples is linguoid. On this basis, an empirical unsteady flow model has been developed and tested with flume experiments in order to predict ripple development in natural depositional environments. The model includes the development of washed-out ripples and upper stage plane bed. The unsteady flow model explains the development and preservation of small scale bedforms in various tidal environments more accurately than previous models. Such bedforms can serve, therefore, as indicators of prevailing hydrodynamic conditions.     

Deposition of climbing-ripple beds: a flume simulation

Thirteen runs were made in a small recirculating flume to simulate the deposition of the climbing-ripple sequences commonly present in fine-grained facies of fluvial and deltaic deposits. These sequences consist of intergradational climbing-ripple cross laminae and draped laminae. The experiments were based on the assumption that stratification type depends mainly on near-bottom flow structure and uniform sediment fallout from an overloaded flow. Various combinations of curves of velocity versus time and of sediment feed versus time in runs lasting from 45 to 840 min were used in an exploratory program; conditions for each run were selected on the basis of experience in previous runs. The runs verified that Type A (erosional-stoss) climbing ripples are produced by aggradation rates that are small relative to ripple migration rate, and Type B (depositional-stoss) climbing ripples are produced by aggradation rates that are large relative to ripple migration rate. Draped lamination results from continued fallout of sediment from suspension after ripple migration ceases or almost ceases. Comparison of geometric details of the ripple stratification produced in the flume runs with that in natural sequences, supplemented by considerations on maximum and minimum migration rates of ripples, suggests times of no more than a few tens of hours for the deposition of the climbing-ripple portions of sequences 10-20 cm thick. Runs in which deposition of a 20 cm sequence took more than 10 h produced such atypical features of ripple geometry as sharp crests, planar lee-side laminae, and angular toeset-foreset contacts.     

Emplacement of massive deposits by sheet flow

Turbidity current and coastal storm deposits are commonly characterized by a basal sandy massive (structureless) unit overlying an erosional surface and underlying a parallel or cross-laminated unit. Similar sequences have been recently identified in fluvial settings as well. Notwithstanding field, laboratory and numerical studies, the mechanisms for emplacement of these massive basal units are still under debate. It is well accepted that the sequence considered here can be deposited by waning-energy flows, and that the parallel-laminated units are deposited under transport conditions corresponding to upper plane bed at the dune–antidune transition. Thus, transport conditions that are more intense than those at the dune–antidune transition should deposit massive units. This study presents experimental, open-channel flow results showing that sandy massive units can be the result of gradual deposition from a thick bedload layer of colliding grains called sheet flow layer. When this layer forms with relatively coarse sand, the non-dimensional bed shear stress associated with skin friction, the Shields number, is larger than a threshold value approximately equal to 0·4. For values of the Shields number smaller than 0·4 the sheet flow layer disappeared, sediment was transported by a standard bedload layer one or two grain diameters thick, and the bed configuration was characterized by downstream migrating antidunes and washed out dunes. Parallel laminae were found in deposits emplaced with standard bedload transport demonstrating that the same dilute flow can gradually deposit the basal and the parallel-laminated unit in presence of traction at the depositional boundary. Further, the experiments suggested that two different types of upper plane bed conditions can be defined, one associated with standard bedload transport at the dune–antidune transition, and the other associated with bedload transport in sheet flow mode at the transition between upstream and downstream migrating antidunes.