Velocity fields and particle trajectories for bed load over subaqueous barchan dunes
Pith reviewed 2026-05-24 18:20 UTC · model grok-4.3
The pith
Bed load grains travel distances up to five times larger over subaqueous barchan dunes than over plane beds while maintaining the same mean velocity.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The central claim is that grains moving over subaqueous barchan dunes exhibit the same mean velocity as those over a plane bed but travel distances up to 5 times larger with higher densities of moving grains. Velocity fields vary along the dune with higher values near the centroid, ranging from 1 to 10 percent of the fluid's cross-sectional mean velocity. Average trajectories show mean displacements of 30 to 60 grain diameters at characteristic velocities of 10 to 20 percent of the fluid velocity, with displacement times between 30 and 90 percent of the settling time showing two asymptotic behaviors near and far from bed load inception.
What carries the argument
High-speed camera imaging combined with custom image processing code to identify and track moving grains, yielding their velocity fields and trajectories over the dune.
If this is right
- Grain velocities vary along the dune, reaching higher values near the centroid.
- Mean grain displacements are 30 to 60 diameters with velocities 10-20% of fluid speed.
- Displacement times show asymptotic behaviors near and far from bed load inception.
- Higher densities of moving grains occur over the dunes compared to plane beds.
Where Pith is reading between the lines
- This suggests that the curved shape of barchan dunes allows grains to be carried farther before settling.
- Models of bed load transport may need adjustment when dunes are present to account for increased path lengths.
- Further tests could examine whether this effect scales with dune size or flow speed.
Load-bearing premise
The image processing correctly identifies and tracks only the moving grains without significant errors from turbulence, overlaps, or lighting.
What would settle it
A repeat experiment showing that average grain travel distances over barchan dunes are not substantially different from those over flat beds would falsify the comparison result.
read the original abstract
This paper presents an experimental investigation of moving grains over subaqueous barchan dunes that consisted of spherical glass beads of known granulometry. Prior to each test run, a pre-determined quantity of grains was poured inside a closed conduit, and the grains settled on its bottom wall forming one conical heap. As different turbulent water flows were imposed, each heap evolved to a barchan dune, which was filmed with a high-speed camera. An image processing code was written to identify some of the moving grains and compute their velocity fields and trajectories. Our results show that the velocity of grains varies along the barchan dune, with higher velocities occurring close to the dune centroid, and that grains trajectories are intermittent. Depending on the region over the dune, we found that the velocity fields present values within 1 and 10 % of the cross-sectional mean velocity of the fluid. Considering the average trajectories of grains moving over a given dune, their mean displacement varies within 30 and 60 grain diameters and their characteristic velocities within 10 and 20 % of the cross-sectional mean velocity of the fluid. The displacement time varies between 30 and 90 % of the settling time, and it seems to have two asymptotic behaviors: one close to bed load inception and other far from it. When compared with bed load over a plane bed, we observe that grains have the same mean velocity, but they travel distances up to 5 times larger, with higher densities of moving grains.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports an experimental investigation of bed-load grain motion over subaqueous barchan dunes formed from spherical glass beads in a closed conduit. High-speed imaging and a custom image-processing code are used to extract velocity fields and trajectories; the authors find that grain velocities vary spatially (higher near the dune centroid), trajectories are intermittent, mean grain velocities match those over plane beds, but travel distances are up to 5 times larger and moving-grain densities are higher.
Significance. If the tracking pipeline is shown to be accurate, the quantitative comparison supplies new data on how bedform topography modulates particle hop lengths at constant mean velocity. This is relevant to sediment-transport and dune-migration models in both fluvial and submarine settings; the use of monodisperse beads and controlled conduit flow is a methodological strength.
major comments (1)
- [Abstract (and implied Methods)] The manuscript provides no description of the image-processing pipeline (identification of moving grains, handling of overlaps, turbulence-induced false positives, or any validation against known velocities or synthetic images). Because the central claim—identical mean velocity but up to 5× longer displacements and higher densities versus plane-bed bedload—rests entirely on the fidelity of these trajectories, the absence of methods, error analysis, and data-exclusion criteria is load-bearing.
minor comments (1)
- [Abstract] The phrasing 'varies within 30 and 60 grain diameters' and 'within 1 and 10 %' is ambiguous; standard usage would be 'between 30 and 60' and 'between 1 and 10 %'.
Simulated Author's Rebuttal
We thank the referee for their thorough review and for highlighting the importance of methodological transparency. The central concern is the absence of a detailed description of the image-processing pipeline. We agree this is a substantive gap that must be addressed and will revise the manuscript accordingly.
read point-by-point responses
-
Referee: [Abstract (and implied Methods)] The manuscript provides no description of the image-processing pipeline (identification of moving grains, handling of overlaps, turbulence-induced false positives, or any validation against known velocities or synthetic images). Because the central claim—identical mean velocity but up to 5× longer displacements and higher densities versus plane-bed bedload—rests entirely on the fidelity of these trajectories, the absence of methods, error analysis, and data-exclusion criteria is load-bearing.
Authors: We agree that the current manuscript lacks a sufficient description of the image-processing methods. In the revised version we will add a dedicated Methods section that details: (i) the criteria and algorithms used to identify moving grains versus stationary ones, (ii) the procedure for handling particle overlaps and occlusions, (iii) steps taken to reduce turbulence-induced false positives (e.g., intensity and size filters, temporal consistency checks), (iv) validation of the pipeline against both synthetic images with known velocities and manual tracking of a subset of frames, (v) quantitative error analysis (position and velocity uncertainties), and (vi) explicit data-exclusion criteria. These additions will allow readers to assess the robustness of the reported velocity fields, trajectories, and comparisons with plane-bed bedload. revision: yes
Circularity Check
No significant circularity
full rationale
This is a purely experimental paper describing high-speed imaging of grain motion over evolving barchan dunes, followed by direct reporting of measured velocities, trajectories, and comparisons to plane-bed bedload. No derivations, model equations, fitted parameters presented as predictions, or self-citation chains appear in the abstract or described methods. The central claims are quantitative observations extracted from image data; they do not reduce to any internal definition or prior self-referential result by construction.
Axiom & Free-Parameter Ledger
Reference graph
Works this paper leans on
-
[1]
Bagnold, The physics of blown sand and desert dunes (Chap- man and Hall, 1941)
R.A. Bagnold, The physics of blown sand and desert dunes (Chap- man and Hall, 1941)
work page 1941
-
[2]
Raudkivi, Loose boundary hydraulics , 1st edn
A.J. Raudkivi, Loose boundary hydraulics , 1st edn. (Pergamon Press, 1976)
work page 1976
-
[3]
Y alin, Mechanics of sediment transport , 1st edn
M.S. Y alin, Mechanics of sediment transport , 1st edn. (Pergamon Press, 1977)
work page 1977
- [4]
- [5]
- [6]
- [7]
- [8]
- [9]
-
[10]
C. Sauermann, P . Rognon, A. Poliakov, H.J. Herrmann, Geo mor- phology 36, 47 (2000)
work page 2000
- [11]
- [12]
-
[13]
Sauermann, Modeling of wind blown sand and desert dune s
G. Sauermann, Modeling of wind blown sand and desert dune s. Ph.D. thesis, Universit¨ at Stuttgart (2001)
work page 2001
- [14]
-
[15]
K. Kroy, G. Sauermann, H.J. Herrmann, Phys. Rev. Lett. 88, 054301 (2002)
work page 2002
- [16]
-
[17]
T. P¨ ahtz, J.F. Kok, E.J.R. Parteli, H.J. Herrmann, Phys . Rev. Lett. 111, 218002 (2013)
work page 2013
-
[18]
G. Seizilles, E. Lajeunesse, O. Devauchelle, M. Bak, Phy s. Fluids 26(1), 013302 (2014)
work page 2014
-
[19]
E. Lajeunesse, L. Malverti, F. Charru, J. Geophys. Res. 115(F4), F04001 (2010)
work page 2010
- [20]
-
[21]
K. Kroy, G. Sauermann, H.J. Herrmann, Phys. Rev. E 66(031302) (2002)
work page 2002
- [22]
- [23]
- [24]
- [25]
- [26]
-
[27]
See Supplementary Material for additional graphics for the re- maining experimental data and a movie showing the motion of grains over a barchan dune
-
[28]
E.M. Franklin, F.T. Figueiredo, E.S. Rosa, J. Braz. Soc. Mech. Sci. Eng. 36, 725 (2014)
work page 2014
- [29]
-
[30]
F.D. C´ u˜ nez, , G.V .G. Oliveira, E.M. Franklin, J. Braz.Soc. Mech. Sci. Eng. 40, 138 (2018)
work page 2018
- [31]
- [32]
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.