REVIEW 4 major objections 4 minor 26 references
Unified Flow Rule of Undeveloped and Fully Developed Dense Granular Flows Down Rough Inclines
T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash
Pith's one-line read The paper claims that a single formula, v = v∞[1 − exp(−l/L)]^(1/2), describes the free-surface velocity of dense granular flows down rough inclines from release to full development, with v∞ and L set only by the dynamic friction…
desk verdict Useful empirical flow rule with fitted exponents; the early-stage rigid-block assumption is the main soft spot before calling it unified. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central machinery is the spatial saturation function $v = v_\infty[1 - \exp(-l/L)]^{1/2}$, obtained by merging the undeveloped velocity law $v_0 = c_0 \sqrt{l(\tan\theta - \mu_r)}$ with the fully developed velocity $v_\infty = c_\infty \mu_r^{3/2}[(\tan\theta - \mu_r)h]^{4/3}$. The development length $L = l v_\infty^2 / v_0^2$ follows algebraically from the two limits. The dimensional analysis for $v_\infty$ borrows from sediment transport: the steady-state velocity is a power law of the excess shear stress $(\tan\theta - \mu_r)h$ times a dimensionless environment scale $\mu_r$, with exponents $p = 3/2$ and $q = 4/3$ fixed by maximizing the Kendall rank correlation across the data. This machinery replaces the old $h/h_s$ scaling with a rule that contains only one material parameter, $\mu_r$, and explicitly accounts for the finite chute length.
What would settle it
Measure the free-surface velocity at several downstream positions along a chute long enough to include both undeveloped and fully developed states, for fixed material, thickness, and angle. If $v^2$ plotted against $l$ does not follow $v_\infty^2[1 - \exp(-l/L)]$ with $L = (c_\infty^2/c_0^2) \mu_r^3[(\tan\theta - \mu_r)h]^{5/3} h$, or if the fitted $L$ does not scale as $[(\tan\theta - \mu_r)h]^{5/3} h$, the rule is falsified. A more direct test of the weakest assumption is to image the velocity profile across the depth early in the flow: if the surface moves faster than the base, the rigid-block law is invalid.
Extended reading notes
Core claim
The central claim is that the free-surface velocity $v$ of a dense granular flow down a rough incline, as a function of travel distance $l$, follows a square-root exponential saturation: $v = v_\infty[1 - \exp(-l/L)]^{1/2}$, where the fully developed velocity is $v_\infty = c_\infty \mu_r^{3/2}[(\tan\theta - \mu_r)h]^{4/3}$ and the development length is $L = (c_\infty^2/c_0^2) \mu_r^3[(\tan\theta - \mu_r)h]^{5/3} h$. Here $\mu_r$ is the dynamic friction coefficient, $h$ is the flow thickness, $\theta$ is the inclination angle, and $c_0$ and $c_\infty$ are proportionality constants that take one set of values for nonmixtures and another for shape-bidisperse mixtures. The rule merges an undeveloped-stage velocity $v_0 = c_0 \sqrt{l(\tan\theta - \mu_r)}$, the law of a rigid block sliding with friction $\mu_r$, with a fully developed velocity that depends only on the excess shear stress and the environment scale $\mu_r$, and not on the stop thickness $h_s(\theta)$. The paper shows that this one formula collapses combined measurements of spheres, two sands, three spheres-sand mixtures, and a zirconia mixture equally well in the undeveloped and fully developed states.
Load-bearing premise
The load-bearing premise is that during the undeveloped stage the entire flow accelerates like a rigid block sliding on the incline with friction coefficient $\mu_r$, giving $v_0 = c_0 \sqrt{l(\tan\theta - \mu_r)}$; if internal shear and basal slip develop early, this law fails and the predicted development length $L$ loses its foundation.
Editorial extensions
If this is right
- For short chutes or thick flows, where the development length $L$ exceeds the travel length $l$, the flow speed stays below the fully developed value; the rule provides a measurements-based estimate of $L$ for the first time.
- The Froude number $v/\sqrt{gh}$ is predicted to be controlled by $\mu_r$, the excess shear stress, and the ratio $l/L$, not by the stop thickness ratio $h/h_s$.
- The stop thickness $h_s(\theta)$ and the second friction coefficient $\mu_2$ drop out of the velocity prediction, so measuring only the dynamic friction coefficient $\mu_r$ is sufficient.
- Compared with the $\mu(I)$-rheology estimate $L \propto (\tan\theta - \mu_r)h^3$, the data imply a substantially stronger dependence on $\tan\theta - \mu_r$ and on $\mu_r$, indicating that the standard rheology misses observed trends.
Reading between the lines
- Because the constants $c_0$ and $c_\infty$ differ between nonmixtures and mixtures, an untested extension is whether they vary continuously with mixing ratio or grain-shape distribution; the paper leaves this open.
- If the null hypothesis that no phase transition occurs at $\mu_2$ holds, the flow rule should remain valid above the stop-thickness divergence, which could be tested in long chutes or with simulations at higher inclination angles.
- The rigid-block assumption implies a testable prediction for the undeveloped stage: the surface velocity should scale as $\sqrt{l}$ with a prefactor independent of $h$ and $\theta$ at fixed $\mu_r$, which high-speed imaging of the velocity profile could confirm or refute.
- Connecting $L$ to natural avalanche runout gives an independent geological check: long-runout avalanches should show spatial velocity profiles that approach the exponential saturation, not linear-in-distance acceleration.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports new laboratory chute measurements of the free-surface velocity of dense granular flows of spheres, two natural sands, spheres-sand mixtures, and a zirconia-based mixture down rough inclines. It argues that the standard flow rules, in which the Froude number scales with h/h_s or (tanθ/μ_r)^2 h/h_s, fail because they are not universal across granular materials and because many measurements are made before the flow is fully developed. The authors propose a unified flow rule v = v∞ [1 − exp(−l/L)]^{1/2}, with v∞ = c∞ μ_r^{3/2} [(tanθ − μ_r) h]^{4/3} and L = (c∞^2/c0^2) μ_r^3 [(tanθ − μ_r) h]^{5/3} h, derived from a dimensional-analysis argument motivated by a recent unification of aeolian and fluvial sediment transport. The rule is compared with the new data and with data from Ref. [6] and is reported to collapse the combined measurements for undeveloped and fully developed flows.
Significance. If the proposed rule holds, it would be a substantial step forward for dense granular chute flows: it replaces the two-parameter h/h_s-based flow rules with a rule built around the single material parameter μ_r, it provides a first measurements-based estimate of the development length L, and it explicitly addresses the pre-saturation regime that previous flow rules ignored. The paper also supplies new experimental data and a simple, falsifiable functional form that can be tested independently. The main strengths are the breadth of tested materials and the direct comparison with earlier experiments. However, the claimed universality is presently only partial: the exponents p and q in Eq. (6) are determined by fitting to the same combined data used to demonstrate the collapse, and the constants c∞ and c0 are fitted separately for nonmixtures and mixtures. The central physical assumption for the undeveloped stage, Eq. (5), is also not directly validated against the multi-position measurements that the paper itself reports.
major comments (4)
- [Eq. (5)] Equation (5), v0 = c0 sqrt(l(tanθ − μ_r)), is the load-bearing assumption for the undeveloped stage, but it is not tested against the paper's own multi-position measurements. Treating the entire undeveloped flow as a rigid block sliding with kinetic friction μ_r is a strong assertion; dense granular flows generally develop internal velocity profiles and basal slip distributions, and the free-surface velocity need not follow a single sliding-friction law. Because Eq. (5) enters directly into Eq. (8) through L = l v∞^2 / v0^2, the predicted development length and the spatial saturation form are not grounded if Eq. (5) fails. I ask the authors to validate Eq. (5) directly using their measurements at l = 0.95 m, 1.3 m, and 1.7 m, or to justify the rigid-block approximation with additional evidence.
- [Eq. (6) and Fig. 5] The exponents p ≈ 3/2 and q ≈ 4/3 in Eq. (6) are determined by weighted-least-squares fitting and Kendall rank maximization using the same combined data that are then displayed as a data collapse in Fig. 5, and c∞ and c0 are fitted separately for nonmixtures and mixtures. Consequently, the collapse in Fig. 5 is not an independent test of the functional form; it is a demonstration that a two-exponent, two-constant family can parameterize the data. The claim of universality would be strengthened substantially by validating Eq. (6) on a holdout subset of the data or against an independent data set, rather than by showing agreement with the same data used for calibration.
- [Eq. (7) and preceding derivation] The translation from the temporal saturation observed in previous DEM simulations (Ref. [18]) to the spatial saturation form in Eq. (7) is not derived rigorously. The step 'l = ∫ v0(t′) dt′ ≈ v0 t/2 and L ∝ v∞ T' assumes that v0(t′) is roughly constant during the undeveloped stage, which is inconsistent with the accelerating rigid-block motion described by Eq. (5). No direct evidence is provided that the spatial velocity profile follows [1 − exp(−l/L)]^{1/2}; the authors themselves note that 'other choices are also possible.' Since Eq. (7) is one of the central predictions, this gap in the derivation should be addressed, either by a more careful asymptotic argument or by a direct test against measured spatial profiles.
- [Fig. 5 caption and text after Eq. (8)] The paper acknowledges that c∞ and c0 are different for nonmixtures and mixtures (c∞ = 11, c0 = 1.5 versus c∞ = 15, c0 = 1.7). The proposed rule therefore contains two material-class-dependent free constants, and the universality across granular materials is only partial. The text attributes the difference to segregation in the mixtures, but as written the model does not predict this difference. The authors should either state clearly that the rule is class-specific with separate calibrated constants, or provide a physical prediction for c∞ and c0 in terms of measurable mixture properties.
minor comments (4)
- [Eq. (7) derivation] In the sentence introducing Eq. (7), the expression 'l = ∫ v0(t′)t′' appears to be a typographical error; it should read 'l = ∫ v0(t′) dt′.'
- [Figs. 4 and 5] Figures 4 and 5 do not show measurement uncertainties. Since the quantitative claims rest on the quality of the data collapse, reporting run-to-run scatter or error bars would make the comparison more convincing.
- [Fig. 5 caption] The numerical values c∞ = 11, c0 = 1.5 and c∞ = 15, c0 = 1.7 appear only in the Fig. 5 caption; they should be listed in the main text or in Table I, together with their uncertainties.
- [Reference [6]] Reference [6] is cited as 'Physics Review E'; the correct journal name is 'Physical Review E.'
Circularity Check
Core scaling v∞ and L are fitted to the data they are then used to explain.
-
fitted input called prediction
[Main text, paragraph preceding Eq. (6), after Eq. (5)]
"We determine the exponents as p ≈ 3/2 and q ≈ 4/3 from weighted-least-squares fitting and maximizing the Kendall rank correlation (τ) coefficients between v/v0 and v∞/v0, according to Eq. (4), simultaneously for both the nonmixture and mixture measurements [15], resulting in two separate rough data collapses (Fig. 5)."
The scaling v∞ ∝ μ_r^p[(tanθ−μ_r)h]^q is proposed on dimensional grounds, but the exponents p and q are obtained by fitting to the same combined measurements that are then displayed as the collapse supporting the law. Eq. (6) is therefore a best-fit empirical curve, not an independent prediction, and its agreement with Fig. 5 is by construction rather than a validation of the dimensional argument.
-
fitted input called prediction
[Fig. 5 caption and paragraph after Eq. (8)]
"Equations (7) and (8) agree with the measurements, though the values of c∞ and c0 are different for nonmixtures and mixtures (Fig. 5)."
The development length L in Eq. (8) is computed as L = l v∞²/v0² using Eq. (5) and the fitted Eq. (6), and the prefactors c∞ and c0 are then tuned per data class (nonmixtures versus mixtures) to make the curves match. Hence Eq. (8) is a calibrated relation rather than a predicted scaling, despite being described in the introduction as 'a prediction for L'.
full rationale
The paper transparently reports that the exponents in Eq. (6) and the coefficients in Eq. (8) are fitted to the same experimental data presented in Fig. 5, so the central quantitative content reduces to empirical calibration rather than independent prediction. The saturation form Eq. (7) is openly an assumption ('other choices are also possible'), and Eq. (5) is a rigid-block sliding assumption, but these are modeling assumptions rather than circular steps per se. The self-citation to Ref. [19] is motivational and not load-bearing, and the comparison with alternative v∞ expressions is legitimate model selection. The main circularity is statistical: the exponents, c∞, and c0 are selected using the same dataset that is then used to claim agreement, so the collapse and the 'prediction' of L are partly by construction. This warrants a score of 6, indicating partial circularity, though not full circularity because the exponential-in-l functional form and the specific variable groupings are not entirely predetermined by the fitting procedure.
Assumptions & free parameters
free parameters (6)
- Exponent p in v∞ ∝ μ_r^p [(tanθ−μ_r)h]^q =
≈ 1.5
- Exponent q in v∞ ∝ μ_r^p [(tanθ−μ_r)h]^q =
≈ 1.333
- Proportionality constant c∞ for nonmixtures =
11
- Proportionality constant c∞ for mixtures =
15
- Proportionality constant c0 for nonmixtures =
1.5
- Proportionality constant c0 for mixtures =
1.7
assumptions (4)
- ad hoc to paper v∞ depends only on excess shear stress τ_ex ∼ (tanθ−μ_r)h and an environment scale C_env ∼ μ_r, and is a power law.
- domain assumption For l ≪ L, the flow accelerates like a rigid block with sliding friction μ_r, giving v0 = c0 sqrt(l(tanθ−μ_r)).
- ad hoc to paper The spatial evolution has the saturation form v = v∞[1−exp(−l/L)]^{1/2} and L = l v∞^2/v0^2.
- domain assumption The flow is dense and air drag is negligible, and h is quasisteady along the chute.
Cite this review
Pith. "Pith review of Unified Flow Rule of Undeveloped and Fully Developed Dense Granular Flows Down Rough Inclines." pith.science (2026). https://pith.science/paper/RT7OI6MQ
@misc{pith2026250110631,
author = {Pith},
title = {Pith review of: Unified Flow Rule of Undeveloped and Fully Developed Dense Granular Flows Down Rough Inclines},
year = {2026},
howpublished = {\url{https://pith.science/paper/RT7OI6MQ}},
note = {Machine review of arXiv:2501.10631}
}
abstract
We report on chute measurements of the free-surface velocity $v$ in dense flows of spheres and diverse sands and spheres-sand mixtures down rough inclines. These and previous measurements are inconsistent with standard flow rules, in which the Froude number $v/\sqrt{gh}$ scales linearly with $h/h_s$ or $(\tan\theta/\mu_r)^2h/h_s$, where $\mu_r$ is the dynamic friction coefficient, $h$ the flow thickness, and $h_s(\theta)$ its smallest value that permits a steady, uniform dense flow state at a given inclination angle $\theta$. This is because the characteristic length $L$ a flow needs to fully develop can exceed the chute or travel length $l$ and because neither rule is universal for fully developed flows across granular materials. We use a dimensional analysis motivated by a recent unification of sediment transport to derive a flow rule that solves both problems in accordance with our and previous measurements: $v=v_\infty[1-\exp(-l/L)]^{1/2}$, with $v_\infty\propto\mu_r^{3/2}\left[(\tan\theta-\mu_r)h\right]^{4/3}$ and $L\propto\mu_r^3\left[(\tan\theta-\mu_r)h\right]^{5/3}h$.
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
-
[6]
C. Ancey, Dry granular flows down an inclined channel: Experimental investigations on the frictional-collisional regime, Physical Review E 65, 011304 (2001)
work page 2001
-
[18]
T. B¨ orzs¨ onyi and R. E. Ecke, Rapid granular flows on a rough incline: Phase diagram, gas transition, and effects of air drag, Physics Review E 74, 061301 (2006)
work page 2006
-
[1]
(5) and (6), L is given by L = c2 ∞/c2 0 µ3 r [(tan θ − µr)h]5/3 h
Using Eqs. (5) and (6), L is given by L = c2 ∞/c2 0 µ3 r [(tan θ − µr)h]5/3 h. (8) Equations (7) and (8) agree with the measurements, though the values of c∞ and c0 are different for non- mixtures and mixtures (Fig. 5). The flow rule derived in this Letter, given by Eqs. (7) and (8), covers undeveloped and fully developed granu- lar flows down rough incli...
work page 2023
-
[2]
D. Keefer and M. C. Larsen, Assessing landslide hazards, Science 316, 1136 (2007)
work page 2007
-
[3]
R. L. Schuster and L. M. Highland, The Third Hans Cloos Lecture. Urban landslides: socioeconomic impacts and overview of mitigative strategies, Bulletin of Engi- neering Geology and the Environment 66, 1 (2007)
work page 2007
-
[4]
Petley, Global patterns of loss of life from landslides, Geology 40, 927 (2012)
D. Petley, Global patterns of loss of life from landslides, Geology 40, 927 (2012)
work page 2012
-
[5]
Pouliquen, Scaling laws in granular flows down rough inclined planes, Physics of Fluids 11, 542 (1999)
O. Pouliquen, Scaling laws in granular flows down rough inclined planes, Physics of Fluids 11, 542 (1999)
1999
-
[7]
T. B¨ orzs¨ onyi and R. E. Ecke, Flow rule of dense granular flows down a rough incline, Physics Review E 76, 031301 (2007)
work page 2007
Show all 26 references
-
[8]
Campbell, C
C. Campbell, C. Brennen, and R. Sabersky, Flow regimes in inclined open-channel flows of granular materials, Pow- der Technology 41, 77 (1985)
1985
-
[9]
Sadjadpour and C
M. Sadjadpour and C. S. Campbell, Granular chute flow regimes: mass flowrates, flowrate limits and clogging, Ad- vanced Powder Technology 10, 175 (1999)
1999
-
[10]
M. Y. Louge and S. C. Keast, On dense granular flows down flat frictional inclines, Physics of Fluids 13, 1213 (2001)
2001
-
[11]
Goujon, N
C. Goujon, N. Thomas, and B. Dalloz-Dubrujeaud, Monodisperse dry granular flows on inclined planes: Role of roughness, The European Physical Journal E 11, 147 (2003)
2003
-
[12]
F´ elix and N
G. F´ elix and N. Thomas, Relation between dry granular flow regimes and morphology of deposits: formation of lev´ ees in pyroclastic deposits, Earth and Planetary Sci- ence Letters 221, 197 (2004)
2004
-
[13]
Deboeuf, E
S. Deboeuf, E. Lajeunesse, O. Dauchot, and B. An- dreotti, Flow rule, self-channelization, and levees in un- confined granular flows, Physical Review Letters 97, 158303 (2006)
2006
-
[14]
M. Y. Louge, A. Valance, P. Lancelot, R. Delannay, and O. Arti´ eres, Granular flows on a dissipative base, Physical Review E 92, 022204 (2015)
2015
-
[15]
H. M. Beakawi Al-Hashemi and O. S. Baghabra Al- Amoudi, A review on the angle of repose of granular materials, Powder Technology 330, 397 (2018)
2018
-
[16]
Note that the modified flow rule based on Ref
See Supplemental Material at [url] for details about the tested granular materials, angle-of-repose measurements, measurement uncertainties, and dimensional analysis be- hind our proposed expression for v∞; for a description of the statistical analysis used to determine p and ...
-
[17]
L. Jing, C. Y. Kwok, Y. F. Leung, and Y. D. Sobral, Characterization of base roughness for granular chute flows, Physical Review E 94, 052901 (2016)
2016
-
[19]
Parez and E
S. Parez and E. Aharonov, Long runout landslides: a solution from granular mechanics, Frontiers in Physics 3, 80 (2015)
2015
-
[20]
P¨ ahtz and O
T. P¨ ahtz and O. Dur´ an, Unification of aeolian and fluvial sediment transport rate from granular physics, Physical Review Letters 124, 168001 (2020)
2020
-
[21]
Bouzid, A
M. Bouzid, A. Izzet, M. Trulsson, E. Cl´ ement, P. Claudin, and B. Andreotti, Non-local rheology in dense granular flows – Revisiting the concept of fluidity, The European Physics Journal E 38, 125 (2015)
2015
-
[22]
Barker, D
T. Barker, D. G. Schaeffer, P. Bohorquez, and J. M. N. T. Gray, Well-posed and ill-posed behaviour of the µ(I)- rheology for granular flow, Journal of Fluid Mechanics 779, 794 (2015)
2015
-
[23]
J. T. Jenkins, Dense inclined flows of inelastic spheres, Granular Matter 10, 47–52 (2007)
2007
-
[24]
J. T. Jenkins, Dense shearing flows of inelastic disks, Physics of Fluids 18, 103307 (2006). 6
2006
-
[25]
Delannay, M
R. Delannay, M. Louge, P. Richard, N. Taberlet, and A. Valance, Towards a theoretical picture of dense gran- ular flows down inclines, Nature Materials 6, 99 (2007)
2007
-
[26]
Patro, M
S. Patro, M. Prasad, A. Tripathi, P. Kumar, and A. Tri- pathi, Rheology of two-dimensional granular chute flows at high inertial numbers, Physics of Fluids 33, 113321 (2021)
2021
Reviewed August 10, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.