Pith. sign in

REVIEW 4 major objections 4 minor 32 references

Distinct Lifetime Scaling Laws of Turbulent Puff in Duct Flow

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Localized turbulent puffs in square ducts are transient structures whose mean lifetime obeys a square-root law below Re≈1450 and a super-exponential law above it.

desk verdict A solid first DNS study of duct-puff lifetime scalings with a plausible but under-validated noisy saddle-node explanation; deserves review once error bars and extrapolation checks are added. read the letter →

arxiv 2507.01583 v1 pith:INCKYLVV submitted 2025-07-02 physics.flu-dyn

Jiashun Guan , Jianjun Tao This is my paper · ORCID
classification physics.flu-dyn PACS 47.27.Cn
keywords turbulentpuffductflowlifetimescalingsaddle-nodebifurcationrelaminarizationReynolds-Orrequationpatternpreservationtransientturbulence
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks whether turbulent puffs in square-duct flow are self-sustained or transient, and what controls their lifetimes. Using direct numerical simulations and the Reynolds-Orr energy equation, it shows that mean puff lifetime follows a square-root law below Re≈1450 and a super-exponential (double-log linear) law above it. The two regimes are explained by reducing the energy dynamics to a noisy saddle-node bifurcation: deterministic critical slowing down below the critical Reynolds number, stochastic barrier crossing above it. The authors conclude that duct puffs never self-sustain in this range and that the decay mechanism is the same as in pipe and channel flows despite the duct's secondary flows.

What carries the argument

The load-bearing object is the noisy saddle-node reduction of the Reynolds-Orr equation. With deterministic production and dissipation written as P_d(E_k) and D_d(E_k), the deterministic growth rate σ_d(E_k)=P_d(E_k)-D_d(E_k)/Re is expanded around the critical energy E_kc where dσ_d/dE_k=0. This yields d(E_k-E_kc)/dt ≈ D_d(E_kc)(Re-Re_c)/$Re_c^{2}$ - (A/2)(E_k-E_kc)^2 + σ_s, the standard stochastic saddle-node normal form. The pattern preservation approximation—that P_d and D_d are functions of E_k alone, nearly independent of Re—is what turns a PDE problem into this one-dimensional normal form and allows the two lifetime scalings to be derived.

What would settle it

Run DNS at Re=1420, 1430, 1440, 1450, 1460, and 1470 with at least 100 independent initial conditions each, and plot both $τ^{-2}$ and ln(ln τ) against Re; the square-root law requires $τ^{-2}$ to be strictly linear in Re below Re_c, and the super-exponential law requires ln(ln τ) to be linear above Re_c. A visible bend or curvature at either end, or a shift of the transition away from 1450 by more than the data scatter, would falsify the claim. Alternatively, compute P_d(E_k) and D_d(E_k) separately for puffs initialized at Re=1500 and Re=1510 and check whether the curves coincide after rescaling by E_k; if they do not, the pattern preservation approximation fails.

Watch

Extended reading notes

Core claim

The central claim is that localized turbulent puffs in a square duct are transient structures whose mean lifetime τ obeys τ=(a1 Re+b1)^(-1/2) for Re<1440, implying a divergence at Re_c=-b1/a1=1450, and ln[ln(τ)]=a2 Re+b2 for Re>1450, a super-exponential growth. These scalings are reproduced by a stochastic saddle-node normal form obtained from the Reynolds-Orr equation under a pattern preservation approximation in which the ensemble-averaged production P_d and dissipation D_d depend only on the disturbance kinetic energy E_k. The critical Reynolds number predicted from the ratio D_d/P_d, namely 1445, closely matches the lifetime-scaling value 1450. Hence the puff's fate is decided by the same noisy saddle-node mechanism previously found in pipe and channel flows: subcritical puffs decay through critical slowing down, while supercritical puffs are metastable and decay by fluctuation-activated barrier crossing.

Load-bearing premise

The pattern preservation approximation—that the ensemble-averaged production and dissipation of disturbance kinetic energy depend only on the current level of disturbance energy and not on Reynolds number or on the puff's history—is the load-bearing premise, because if P_d and D_d vary with Re or with the puff's internal dynamics beyond E_k, the quadratic expansion and the two scaling laws do not follow.

Editorial extensions

If this is right

  • If correct, duct puffs are transient for all Re covered here; there is no self-sustained puff below Re≈1450, only a metastable one above it.
  • The square-root scaling below Re_c follows from critical slowing down in the ghost of the saddle-node, so lifetime data can be used to locate Re_c independently of the DNS fit.
  • The super-exponential regime implies that above Re_c the mean lifetime grows extremely fast with Re, so laboratory or DNS experiments must use very long runs to observe decay.
  • The same mechanism as in pipe and channel flow suggests a universal description of localized turbulence decay in wall-bounded shear flows, despite structural differences like corner vortices.
  • The predicted Re_c≈1445 from D_d/P_d can be cross-checked against lifetime measurements, providing a direct test of the theory.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct testable extension: measure the full lifetime distribution, not just the mean. The saddle-node model predicts a specific non-exponential distribution below Re_c and an exponential, memoryless distribution above Re_c; a mismatch would rule out the reduction.
  • The pattern preservation approximation may break down at higher Re where puff splitting occurs; then the single-variable normal form would need extra dimensions, possibly changing the scaling laws.
  • One could check whether the same two-regime lifetime scaling appears for rectangular ducts of different aspect ratios, using the aspect ratio as a continuous parameter to see how Re_c shifts and whether the universal form survives.
  • If the critical energy E_kc≈0.355 corresponds to an edge state, the theory connects directly to the dynamical-systems view of the laminar-turbulent boundary; verifying that the unstable saddle branch is an edge state would link lifetime scaling to the edge manifold.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. This manuscript reports direct numerical simulations of localized turbulent puffs in square duct flow over a Reynolds number range around 1400–1600. The central claim is that puffs are always transient, with the mean lifetime τ following τ = (a1 Re + b1)^{-1/2} for Re < 1440 and ln[ln(τ)] = a2 Re + b2 for Re > 1450, with a critical Reynolds number Re_c ≈ 1450 separating the two regimes. The authors propose a theoretical explanation based on a pattern-preservation approximation, which reduces the Reynolds–Orr kinetic energy equation to a noisy saddle-node bifurcation equation (Eq. 9). They argue that the square-root law arises from critical slowing down in the subcritical regime and that the super-exponential law arises from noise-activated barrier crossing in the supercritical regime. They also document ensemble-averaged secondary flow structures and conclude that the decay mechanism is geometrically invariant between pipe and duct flows.

Significance. If the claims hold, this is a valuable contribution: it extends the transient-puff picture and the noisy saddle-node framework from pipe flow to duct flow, provides quantitative lifetime scaling laws, and is supported by a substantial DNS campaign of about 1000 individual puff decay simulations. The paper is clearly written, the numerical setup is documented, and the theoretical reduction is conceptually appealing. The main caveat is that the central scaling laws are fits to mean lifetimes without reported uncertainties, and the pattern-preservation collapse that underpins the theory is established only visually from the same DNS data. The significance is therefore conditional on these statistical and methodological gaps being addressed.

major comments (4)
  1. [Fig. 3, Eqs. (4)–(5)] The two central scaling laws are fitted to mean lifetimes without any reported uncertainty. With M = 100 and M = 50 per Reynolds number and approximately exponential lifetime distributions, the standard error of each mean is on the order of τ/√M (10–14%), which is large enough to affect the fitted exponents and the inferred Re_c. Please provide bootstrap confidence intervals for the fitted parameters, show the lifetime distributions for representative Re, and demonstrate that the choice of relaminarization threshold E_k = 0.05 does not change the fitted scaling laws quantitatively.
  2. [Fig. 4, text before Eq. (7)] The pattern preservation approximation is the load-bearing assumption, but the collapse of P_d and D_d curves across Re is only visual. The text states that at low Re puff energies 'seldom reach high values', so the high-E_k portion of the curves, where E_kc = 0.355 is located, is sparsely sampled. The quintic fits and the predicted Re_c = 1445 from Eq. (7) may therefore be artifacts of extrapolation. Please quantify the data density in E_k, add scatter or error bars to Fig. 4, test the sensitivity of Re_c and E_kc to the polynomial order and to the fitted E_k range, and show that P_d and D_d are statistically indistinguishable across Re. If P_d and D_d depend on Re or on puff history beyond E_k, the quadratic expansion in Eq. (8) and the derived scaling laws do not follow.
  3. [Eq. (4), Eq. (5), Fig. 3] The critical Reynolds number Re_c = 1450 is obtained from the square-root fit (Eq. 4) using data with Re < 1440, and this same Re_c is then used to partition the data into the two scaling regimes, so the boundary is not determined independently. This circularity makes it difficult to evaluate whether the two-regime description is actually required. Please fit both regimes jointly with the breakpoint as a free parameter, or determine Re_c independently from the crossing of the two fits, and report the uncertainty in Re_c. The value Re_c = 1445 from Eq. (7) is a consistency check, not an independent determination.
  4. [Eq. (9) and following] The derivation of the super-exponential scaling ln[ln τ] = a2 Re + b2 from the noisy saddle-node equation is asserted through a reference to barrier-crossing theory, rather than derived or quantitatively matched. To support the mechanism claim, the authors should either derive the scaling for Eq. (9) with the noise amplitude estimated from DNS, or compare the predicted barrier height and escape rate with the fitted a2 and b2. As written, the empirical Eq. (5) and the theoretical Eq. (9) are connected only qualitatively.
minor comments (4)
  1. [Abstract and text] There are several typographical errors, including 'Naiver-Stokes' for 'Navier-Stokes' and missing spaces in phrases such as 'analyzethe' and 'and 𝑅𝑒𝑐'.
  2. [Reference [25]] The reference for Nek5000 lists 'F. Paul, J. W. L. Fischer, S. G. Kerkemeier'; the standard citation is P. F. Fischer, J. W. Lottes, and S. G. Kerkemeier. Please correct the author list.
  3. [Fig. 3] The ordinate labels should explicitly state the transformed variables (τ^{-2} and ln[ln τ]) and should include error bars or a statement explaining why they are omitted. The legend would benefit from clarifying the meaning of Re_init = 1500 and 1510.
  4. [Data Availability] The statement that the data are 'available within the article' is not accurate for the full lifetime dataset; please deposit the individual lifetimes and the P_d and D_d curves in a public repository so that the fits can be reproduced.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the theoretical reduction is an internal consistency check with independent observables, and the self-citations are not load-bearing.

full rationale

I walked the derivation chain from the Reynolds-Orr equation (Eq. 3) through the pattern-preservation reduction (Eqs. 6-9). The lifetime scaling laws, Eqs. (4) and (5), are empirical fits to DNS decay times, and the model is not used to derive their fitted coefficients; it only supplies the functional forms via the noisy saddle-node scenario. The critical Reynolds number from Eq. (7), Rec=1445, is obtained from quintic fits to DNS-derived production and dissipation curves and compared with Rec=1450 from the lifetime fit. This is an internal cross-consistency check between two distinct observables (energy budget vs. decay time), not a reduction of the prediction to its own fitted input: the agreement is nontrivial. The pattern-preservation approximation is not imported solely by self-citation; it is explicitly checked in Fig. 4, and the paper notes the low-Re sampling limitation at high E_k, which is a robustness concern rather than circularity. Self-citations to refs. 24 and 29 supply precedent and vocabulary, but the saddle-node reduction is rederived here, so these citations are not load-bearing. The main weakness is the sparsity of high-E_k trajectories at low Re, which affects the reliability of the extrapolated critical point but does not make the derivation circular.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central scaling laws are anchored by fitted parameters a1, b1, a2, and b2 and by a relaminarization threshold. The theoretical reduction additionally relies on quintic fits to ensemble-averaged production and dissipation, which are used to compute Rec=1445. No new physical entities are introduced.

free parameters (6)
  • a1 (square-root scaling prefactor) = -3.08e-6
    Fitted to mean lifetime data for Re<1440 in Eq. (4).
  • b1 (square-root scaling offset) = 4.46e-3
    Fitted with a1 in Eq. (4); determines Rec=-b1/a1=1450.
  • a2 (super-exponential slope) = 4.94e-3
    Fitted to ln[ln(tau)] versus Re for Re>1450 in Eq. (5).
  • b2 (super-exponential intercept) = -5.48
    Fitted with a2 in Eq. (5).
  • quintic polynomial coefficients for P_d(E_k) and D_d(E_k) = not reported
    Ensemble-averaged production and dissipation curves in Fig. 4 are fitted with quintic polynomials; these fits are used to compute Rec=1445 via Eq. (7) and the expansion coefficients in Eq. (8).
  • relaminarization threshold E_k=0.05 = 0.05
    Chosen criterion for puff death; authors assert insensitivity but do not show the test.
assumptions (5)
  • domain assumption Pattern preservation approximation: P_d and D_d depend only on E_k and are nearly independent of Re.
    Invoked before Eq. (7) and validated by Fig. 4; load-bearing for reducing Reynolds-Orr to a one-dimensional noisy saddle-node equation.
  • domain assumption The deterministic components P_d and D_d are well defined as ensemble averages over E_k slices of width 0.02.
    Requires that puff trajectories sample the same attractor and that averaging over finite slices represents the deterministic dynamics.
  • standard math Noisy saddle-node bifurcation theory (critical slowing down and barrier crossing) applies to the reduced equation.
    Uses classical results from Strogatz and Hathcock and Sethna; assumes the noise term can be treated as a weak stochastic perturbation.
  • domain assumption Basic flow is steady and parallel with constant flow rate; DNS resolves relevant scales.
    Validation shows less than 3% difference in mean disturbance kinetic energy at higher resolution, but lifetime statistics may be more sensitive than mean energy.
  • domain assumption Lifetimes at Re>Rec follow a memoryless exponential process enabling mean lifetime to be summarized by a single tau.
    Implied by barrier crossing model; not directly verified by reported lifetime distributions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Distinct Lifetime Scaling Laws of Turbulent Puff in Duct Flow." pith.science (2026). https://pith.science/paper/INCKYLVV

@misc{pith2026250701583,
  author       = {Pith},
  title        = {Pith review of: Distinct Lifetime Scaling Laws of Turbulent Puff in Duct Flow},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/INCKYLVV}},
  note         = {Machine review of arXiv:2507.01583}
}
abstract

The spatio-temporal dynamics of localized turbulent puffs $-$ the characteristic transitional structures in square duct flows $-$ are investigated through direct numerical simulations and theoretical analyses. It is revealed that the turbulent puffs are transient structures, exhibiting distinct relaminarization regimes bifurcated at a critical Reynolds number $Re_c\simeq1450$. Puff's mean lifetimes at the subcritical regime ($Re<Re_c$) follow a square-root scaling law with increasing $Re$, transitioning to a super-exponential scaling in the supercritical regime ($Re > Re_c$). By implementing pattern preservation approximation, the Reynolds-Orr kinetic energy equation is reduced to a noisy saddle-node bifurcation equation, which explains the observed scaling laws in terms of the deterministic decay governed by the critical slowing down at the subcritical regime, and the abrupt decay activated by the stochastic fluctuations. Despite geometric confinement inducing unique secondary flows, e.g., corner-localized streamwise vortex pairs, corner-aligned high-speed streaks, and forked low-speed streaks, the puff lifetime statistics remain analogous to those in pipe flows, suggesting geometric invariance in decay mechanisms for transitional wall-surrounded turbulence.

Figures

Figures reproduced from arXiv: 2507.01583 by the authors.

Figure 1
Figure 1. FIG. 1. Flow field of a turbulent puff obtained at [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The ensemble-averaged disturbance flow field [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Mean lifetime [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Trajectories of puffs in the (a) [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 31 canonical work pages

  1. [1]

    author author O. Reynolds ,\ title title An experimental investigation of the circumstances which determine whether the motion of water shall be direct or sinuous, and of the law of resistance in parallel channels , \ @noop journal journal Philos. Trans. R. Soc. Lond. \ volume 174 ,\ pages 935--982 ( year 1883 ) NoStop

    work page

  2. [2]

    Eckhardt , author T

    author author B. Eckhardt , author T. M. \ Schneider , author B. Hof , \ and\ author J. Westerweel ,\ title title Turbulence transition in pipe flow , \ @noop journal journal Annu. Rev. Fluid Mech. \ volume 39 ,\ pages 447--468 ( year 2007 ) NoStop

    work page 2007

  3. [3]

    Mullin ,\ title title Experimental studies of transition to turbulence in a pipe , \ @noop journal journal Annu

    author author T. Mullin ,\ title title Experimental studies of transition to turbulence in a pipe , \ @noop journal journal Annu. Rev. Fluid Mech. \ volume 43 ,\ pages 1--24 ( year 2011 ) NoStop

    work page 2011

  4. [4]

    Tatsumi \ and\ author T

    author author T. Tatsumi \ and\ author T. Yoshimura ,\ title title Stability of the laminar flow in a rectangular duct , \ @noop journal journal J. Fluid Mech. \ volume 212 ,\ pages 437--449 ( year 1990 ) NoStop

    work page 1990

  5. [5]

    author author K. V. \ Demyanko \ and\ author Y. M. \ Nechepurenko ,\ title title Linear stability analysis of P oiseuille flow in a rectangular duct , \ @noop journal journal Russ. J. Numer. Anal. Math. Modelling \ volume 28 ,\ pages 125--148 ( year 2013 ) NoStop

    work page 2013

  6. [6]

    u ber die Geschwindigkeitsverteilung in turbulenten Str\

    author author J. Nikuradse ,\ title Untersuchungen \" u ber die Geschwindigkeitsverteilung in turbulenten Str\" o mungen ,\ @noop Ph.D. thesis ,\ school G\" o ttingen ( year 1926 ),\ note vDI Forsch. 281 NoStop

    work page 1926

  7. [7]

    Uhlmann , author A

    author author M. Uhlmann , author A. Pinelli , author G. Kawahara , \ and\ author A. Sekimoto ,\ title title Marginally turbulent flow in a square duct , \ @noop journal journal J. Fluid Mech. \ volume 588 ,\ pages 153--162 ( year 2007 ) NoStop

    work page 2007

  8. [8]

    Gavrilakis ,\ title title Numerical simulation of low-reynolds-number turbulent flow through a straight square duct , \ @noop journal journal J

    author author S. Gavrilakis ,\ title title Numerical simulation of low-reynolds-number turbulent flow through a straight square duct , \ @noop journal journal J. Fluid Mech. \ volume 244 ,\ pages 101--129 ( year 1992 ) NoStop

    work page 1992

Show all 32 references
  1. [9]

    Huser \ and\ author S

    author author A. Huser \ and\ author S. Biringen ,\ title title Direct numerical simulation of turbulent flow in a square duct , \ @noop journal journal J. Fluid Mech. \ volume 257 ,\ pages 65 ( year 1993 ) NoStop

    1993

  2. [10]

    Pinelli , author M

    author author A. Pinelli , author M. Uhlmann , author A. Sekimoto , \ and\ author G. Kawahara ,\ title title Reynolds number dependence of mean flow structure in square duct turbulence , \ @noop journal journal J. Fluid Mech. \ volume 644 ,\ pages 107--122 ( year 2010 ) NoStop

    2010

  3. [11]

    Takeishi , author G

    author author K. Takeishi , author G. Kawahara , author H. Wakabayashi , author M. Uhlmann , \ and\ author A. Pinelli ,\ title title Localized turbulence structures in transitional rectangular-duct flow , \ @noop journal journal J. Fluid Mech. \ volume 782 ,\ pages 368--379 ( ...

    2015

  4. [12]

    Barkley , author B

    author author D. Barkley , author B. Song , author V. Mukund , author G. Lemoult , author M. Avila , \ and\ author B. Hof ,\ title title The rise of fully turbulent flow , \ 10.1038/nature15701 journal journal Nature \ volume 526 ,\ pages 550--553 ( year 2015 ) NoStop

    doi

  5. [13]

    Guan \ and\ author J

    author author J. Guan \ and\ author J. Tao ,\ title title Propagation unity of turbulent fronts in polygonal duct flows , \ in\ @noop booktitle International Congress of Theoretical and Applied Mechanics \ ( year 2024 ) NoStop

    2024

  6. [14]

    Khan , author S

    author author H. Khan , author S. Anwer , author N. Hasan , \ and\ author S. Sanghi ,\ title title Laminar to turbulent transition in a finite length square duct subjected to inlet disturbance , \ @noop journal journal Phys. Fluids \ volume 33 ,\ pages 065128 ( year 2021 ) NoStop

    2021

  7. [15]

    author author I. J. \ Wygnanski \ and\ author F. H. \ Champagne ,\ title title On transition in a pipe. part 1. the origin of puffs and slugs and the flow in a turbulent slug , \ @noop journal journal J. Fluid Mech. \ volume 59 ,\ pages 281--335 ( year 1973 ) NoStop

    1973

  8. [16]

    Hof , author J

    author author B. Hof , author J. Westerweel , author T. M. \ Schneider , \ and\ author B. Eckhardt ,\ title title Finite lifetime of turbulence in shear flows , \ @noop journal journal Nature \ volume 443 ,\ pages 59--62 ( year 2006 ) NoStop

    2006

  9. [17]

    Peixinho \ and\ author T

    author author J. Peixinho \ and\ author T. Mullin ,\ title title Decay of turbulence in pipe flow , \ @noop journal journal Phys. Rev. Lett. \ volume 96 ,\ pages 094501 ( year 2006 ) NoStop

    2006

  10. [18]

    Avila , author A

    author author M. Avila , author A. P. \ Willis , \ and\ author B. Hof ,\ title title On the transient nature of localized pipe flow turbulence , \ @noop journal journal J. Fluid Mech. \ volume 646 ,\ pages 127--136 ( year 2010 ) NoStop

    2010

  11. [19]

    Hof , author A

    author author B. Hof , author A. de Lozar , author D. J. \ Kuik , \ and\ author J. Westerweel ,\ title title Repeller or attractor? selecting the dynamical model for the onset of turbulence in pipe flow , \ @noop journal journal Phys. Rev. Lett. \ volume 101 ,\ pages 214501 ( ...

    2008

  12. [20]

    Avila , author D

    author author K. Avila , author D. Moxey , author A. de Lozar , author M. Avila , author D. Barkley , \ and\ author B. Hof ,\ title title The onset of turbulence in pipe flow , \ @noop journal journal Science \ volume 333 ,\ pages 192--196 ( year 2011 ) NoStop

    2011

  13. [21]

    Wygnanski , author M

    author author I. Wygnanski , author M. Sokolov , \ and\ author D. Friedman ,\ title title On transition in a pipe. part 2. the equilibrium puff , \ @noop journal journal J. Fluid Mech. \ volume 69 ,\ pages 283--304 ( year 1975 ) NoStop

    1975

  14. [22]

    Barkley ,\ title title Theoretical perspective on the route to turbulence in a pipe , \ 10.1017/jfm.2016.465 journal journal J

    author author D. Barkley ,\ title title Theoretical perspective on the route to turbulence in a pipe , \ 10.1017/jfm.2016.465 journal journal J. Fluid Mech. \ volume 803 ,\ pages P1 ( year 2016 ) NoStop

    2016 doi

  15. [23]

    author author H. Y. \ Shih , author T. L. \ Hsieh , \ and\ author N. Goldenfeld ,\ title title Ecological collapse and the emergence of travelling waves at the onset of shear turbulence , \ @noop journal journal Nat. Phys. \ volume 12 ,\ pages 245--248 ( year 2016 ) NoStop

    2016

  16. [24]

    Guan \ and\ author J

    author author J. Guan \ and\ author J. Tao ,\ title title The onset of metastable turbulence in pipe flow , \ @noop journal journal https://doi.org/10.48550/arXiv.2504.14465 \ ( year 2025 ) NoStop

    2025 doi arXiv

  17. [25]

    Paul , author J

    author author F. Paul , author J. W. L. \ Fischer , \ and\ author S. G. \ Kerkemeier ,\ @noop title Nek5000 webpage , \ howpublished http://nek5000.mcs.anl.gov ( year 2008 ),\ note [Accessed: 2025-03-06] NoStop

    2008

  18. [26]

    author author A. P. \ Willis \ and\ author R. R. \ Kerswell ,\ title title Critical behavior in the relaminarization of localized turbulence in pipe flow , \ 10.1103/PhysRevLett.98.014501 journal journal Phys. Rev. Lett. \ volume 98 ,\ pages 014501 ( year 2007 ) NoStop

    doi

  19. [27]

    Song , author D

    author author B. Song , author D. Barkley , author B. Hof , \ and\ author M. Avila ,\ title title Speed and structure of turbulent fronts in pipe flow , \ 10.1017/jfm.2017.14 journal journal J. Fluid Mech. \ volume 813 ,\ pages 1045--1059 ( year 2017 ) NoStop

    2017 doi

  20. [28]

    Avila , author F

    author author M. Avila , author F. Mellibovsky , author N. Roland , \ and\ author B. Hof ,\ title title Streamwise-localized solutions at the onset of turbulence in pipe flow , \ @noop journal journal Phys. Rev. Lett. \ volume 110 ,\ pages 224502 ( year 2013 ) NoStop

    2013

  21. [29]

    Zhang \ and\ author J

    author author L. Zhang \ and\ author J. Tao ,\ title title Pattern preservation during the decay and growth of localized wave packet in two-dimensional channel flow , \ @noop journal journal Phys. Fluids \ volume 34 ,\ pages 064110 ( year 2022 ) NoStop

    2022

  22. [30]

    Hathcock \ and\ author J

    author author D. Hathcock \ and\ author J. P. \ Sethna ,\ title title Reaction rates and the noisy saddle-node bifurcation: Renormalization group for barrier crossing , \ @noop journal journal Phys. Rev. Research \ volume 3 ,\ pages 013156 ( year 2021 ) NoStop

    2021

  23. [31]

    Manneville ,\ @noop title Instabilities, Chaos and Turbulence: an Introduction to Nonlinear Dynamics and Complex Systems \ ( publisher Imperial College Press ,\ year 2004 ) NoStop

    author author P. Manneville ,\ @noop title Instabilities, Chaos and Turbulence: an Introduction to Nonlinear Dynamics and Complex Systems \ ( publisher Imperial College Press ,\ year 2004 ) NoStop

    2004

  24. [32]

    author author S. Strogatz ,\ @noop title Nonlinear Dynamics and Chaos: With Applications to Physics, Biology, Chemistry, and Engineering \ ( publisher CRC Press ,\ address Boca Raton London New York ,\ year 2019 ) NoStop

    2019

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.