Resonance-Driven Intermittency and Extreme Events in Turbulent Scalar Transport with a Mean Gradient

Di Qi; Mustafa A Mohamad

REVIEW 1 major objections 6 minor 23 references

Resonance-Driven Intermittency and Extreme Events in Turbulent Scalar Transport with a Mean Gradient

T0 review · 1 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Tracer bursts in turbulent transport with a mean gradient are driven by a resonance between zonal and shear flow phase speeds, not by instabilities; the stationary tracer PDF is a Gaussian mixture peaked at that resonance.

desk verdict Useful closed-form intermittency model with a real resonance mechanism, but the printed multi-mode PDF has a factor-of-2 normalization slip that needs fixing. read the letter →

arxiv 2505.21688 v2 pith:F3GI3S3C submitted 2025-05-27 cs.CE math-phmath.DSmath.MPnlin.CDphysics.flu-dyn

Mustafa A Mohamad , Di Qi This is my paper

classification cs.CEmath-phmath.DSmath.MPnlin.CDphysics.flu-dyn MSC 76F2560H1076M35

keywords turbulentdiffusionpassivescalartracerintermittencyextremeeventsresonanceconditionconditionalGaussianstatisticszonalflowmeangradient

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

This paper asks what creates the intermittent bursts and heavy-tailed statistics observed in passive tracer fields transported by turbulent flows with a mean gradient. The authors construct an analytically tractable stochastic model with a zonal cross-sweep and a shear flow, and derive an explicit formula for the stationary distribution of the tracer. The formula shows that the tracer's conditional variance has a sharp peak when the phase speeds of the zonal and shear flows line up, a resonance condition. As the zonal flow fluctuates across that threshold, the corresponding Fourier modes are excited and the tracer bursts. The paper argues that this makes intermittency a predictable, resonance-crossing effect rather than the product of transient instabilities, and it validates the prediction across nonlinear zonal flows, several shear flow models, and different energy spectra.

What carries the argument

The load-bearing object is the conditional tracer variance $\tilde\Sigma_k(u)=\alpha^2 E_{v,k}/(\gamma_{T,k}^2+\omega_{R,k}(u)^2)$, peaked at the resonance condition $\omega_{R,k}(u)=0$. The argument rests on the multiscale approximation that freezes the slowly varying zonal flow $u$ while the tracer responds, and on the conditional Gaussianity of the shear and tracer modes given a zonal trajectory, which turns the full probability density into an integral (a Gaussian mixture) over the zonal flow's stationary density $p_u(u)$.

What would settle it

In a single-mode simulation, compute the tracer's empirical conditional variance as a function of the zonal flow value $u$; the resonance mechanism predicts a sharp Lorentzian peak at $u=u_{\mathrm{res},k}=-b_k/(a_k+k)$ with the shape $\alpha^2 E_{v,k}/(\gamma_{T,k}^2+\omega_{R,k}(u)^2)$. A clear peak elsewhere, or no peak at all, would falsify the claim.

Watch

Extended reading notes

Core claim

The central claim is that, in the limit where the velocity fields vary much more slowly than the tracer, the stationary conditional variance of the tracer at wavenumber $k$ is $\tilde\Sigma_k(u)=\alpha^2 E_{v,k}/(\gamma_{T,k}^2+\omega_{R,k}(u)^2)$, a Lorentzian function of the instantaneous zonal flow $u$ with a sharp peak where the resonance frequency $\omega_{R,k}(u)=\omega_{T,k}-\omega_{v,k}=-(a_k+k)u-b_k$ vanishes. Each time the fluctuating zonal flow crosses the threshold $u_{\mathrm{res},k}=-b_k/(a_k+k)$, the variance of mode $k$ is amplified and an intermittent burst occurs. Averaging over the zonal flow's invariant density $p_u(u)$ gives the stationary tracer PDF as the Gaussian mixture $p(\lambda)=\int(2\pi\tilde\Sigma(u))^{-1/2}\exp(-\lambda^2/2\tilde\Sigma(u))\,p_u(u)\,du$ with $\tilde\Sigma(u)=\sum_k \alpha^2 E_{v,k}/(\gamma_{T,k}^2+\omega_{R,k}(u)^2)$. The paper presents this as a resonance-driven mechanism for intermittency, distinct from instability-driven intermittency, and supports it with numerical experiments across linear and nonlinear zonal flow models, random, advective, and quasi-geostrophic shear flows, and equipartition and Kolmogorov energy spectra.

Load-bearing premise

The closed-form tracer PDF assumes the velocity fields evolve much more slowly than the tracer, so the zonal flow can be treated as frozen while the tracer responds; the paper's own simulations show this approximation degrades when strong multiplicative noise shortens the zonal flow's timescale.

Editorial extensions

If this is right

The tracer PDF is fully determined by the zonal flow's invariant distribution and its threshold-crossing frequencies, so knowing the flow statistics lets one predict the rate and size of tracer bursts.
Non-Gaussian zonal flows do not automatically produce more intermittent tracers; what matters is how often the flow crosses resonance thresholds, which challenges simple linearization of the zonal flow.
Dispersive shear flows spread resonance thresholds across wavenumbers, giving sequential mode excitation and smoother extremes, while random and non-dispersive flows synchronize thresholds and produce sharper, more intermittent bursts with fine-scale structure.
The energy spectrum of the shear flow sets which scales dominate extreme events: equipartition spectra excite small scales, while Kolmogorov spectra make extremes large-scale dominated.

Reading between the lines

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

The Gaussian-mixture form of the tracer PDF implies the tracer's kurtosis is $3\,\mathbb{E}[\tilde\Sigma(u)^2]/\mathbb{E}[\tilde\Sigma(u)]^2$, which offers a direct diagnostic for when a flow sits in the intermittency regime.
The resonance-crossing mechanism is a special case of a broader principle: any linear system driven by a slowly varying parameter that sweeps its frequency mismatch through zero will develop intermittent heavy tails, so the analysis may transfer to other geophysical or engineering systems besides tracer transport.
A direct numerical test of the mechanism is to measure the tracer's conditional variance conditioned on the zonal flow in a two-dimensional turbulent simulation with a mean gradient; the predicted Lorentzian peak at $u_{\mathrm{res},k}$ should be observable even where the closed-form PDF deviates at finite $\epsilon$.

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.

Desk Editor's Note

read the letter

Quick read: the paper delivers a genuinely useful closed-form PDF for a passive tracer in a stochastic zonal+shear flow model, and the core claim—intermittency as resonance crossing, not instability—is supported by the numerics. The new bit is the explicit stationary mixture with mode-resolved resonance thresholds, extending the Bourlioux-Majda and Majda-Tong line to nonlinear zonal dynamics. I'd send it to peer review.

What it does well: the derivation in Prop 4.2 is a clean quasistatic asymptotic; the numerics cover linear/nonlinear zonal flows, several shear models, and two spectra, and the empirical PDFs broadly track the analytic curves. The paper also flags its own weak spot: strong multiplicative noise (B=2.5) shortens the zonal timescale and degrades the PDF agreement, as reported in Sec. 5.2. That's honest.

Soft spots, in order. First, Theorem 4.3 as printed is off by a factor of two. eΣ_k is the variance of the complex mode bT_k; the real field T(x) sums ±k pairs, so its variance is 2Σ_{k∈N} eΣ_k. The theorem's eΣ(u) sums only positive k, so the mixture is too narrow by √2. The multi-mode figures seem to match, which means either the implementation used the factor of 2 or the plotted quantity is a single complex mode—either way the theorem can't be reproduced from the text. This is a fixable bug, but it's load-bearing for the claimed explicit PDF. Second, the quasistatic approximation is real: it works exactly when the velocity is much slower than the tracer, and the authors show the degradation themselves. That's a well-stated limitation, not a hidden flaw. Third, no code or data are provided, so exact reproduction is impossible; for a computational paper that's a minor but real inconvenience. Also there's a duplicated reference ([11]=[12])—sloppy but harmless.

Bottom line: the resonance mechanism is not new—the paper credits [1] for that—but the closed-form stochastic generalization is. The factor-of-2 issue needs fixing before I'd rely on the quantitative PDFs, but the qualitative picture and the model framework are solid. Worth a serious referee.

Referee Report

1 major / 6 minor

Summary. The paper introduces an analytically tractable stochastic model for passive scalar transport with a mean gradient, coupling a zonal flow (linear OU or nonlinear cubic SDE with CAM noise) to a shear flow with wavelike/dispersive dynamics and a prescribed energy spectrum. Working in Fourier space and exploiting conditional Gaussianity given the zonal flow trajectory, the authors derive a closed-form stationary PDF for the tracer field under a slow-fast timescale separation (ε→0). The central claim is that extreme events and non-Gaussian statistics arise from a resonance condition ω_R,k = 0 between the zonal flow, shear wave, and tracer phase speeds, which produces sharp peaks in the conditional tracer variance. The claim is supported by single- and multi-mode numerical experiments with different energy spectra and zonal flow models, with an acknowledged degradation of the analytical approximation for strong multiplicative noise (B = 2.5).

Significance. If the closed-form PDF and the resonance mechanism hold, this paper provides a valuable testbed for uncertainty quantification and data assimilation of non-Gaussian tracer statistics, and offers a clear physical mechanism (resonance rather than instability) for tracer intermittency. The paper is commendably explicit: it provides exact trajectory solutions (Prop. 3.1), a rigorous conditional Gaussian framework, an honest discussion of the timescale-separation breakdown under strong multiplicative noise, and extensive numerical validation across shear-flow classes, energy spectra, and zonal-flow regimes. However, the central closed-form expression in Theorem 4.3 contains a normalization error (a missing factor of 2 for the real-valued tracer field), which must be corrected and the numerical comparisons re-verified before the result can be considered reproducible.

major comments (1)

[Theorem 4.3, Eq. (59)] The theorem states the stationary tracer PDF as a mixture of Gaussians with variance eΣ(u) = Σ_{k∈N} eΣ_k(u), where Prop. 4.2 defines eΣ_k(u) as the conditional variance of the complex Fourier mode bT_k (E|bT_k|²). Because the physical tracer is real, T(x) = Σ_k bT_k e^{ikx} with bT_{-k} = bT_k*, each ±k pair contributes 2Re(bT_k e^{ikx}) to the real field, so the conditional variance of T(x) is 2 Σ_{k∈N} eΣ_k(u), not Σ_{k∈N} eΣ_k(u). The printed formula therefore predicts a PDF that is too narrow by a factor of √2. This is an algebraic normalization error independent of the ε→0 quasistatic approximation. The authors should correct Eq. (59) (multiply by 2, or explicitly define eΣ(u) as the variance of the real field) and then re-check the multi-mode analytical curves in Sec. 6, which appear to match the numerics; either the implementation used the correct factor 2 (contradicting the printed formula) or the plotted quantity is not the real tracer field T(x). Please clarify which is the case and provide corrected analytical curves.

minor comments (6)

[Sec. 5.2 and Sec. 6.1.2] The text correctly notes that strong multiplicative noise (B = 2.5) shortens the zonal-flow timescale and degrades the timescale separation underlying Prop. 4.2, but Figs. 5 and 8 still overlay the analytical PDF on the empirical PDF without quantifying the mismatch. Since the paper claims validation across regimes, please add a quantitative error measure (e.g., relative entropy or L1 error) for the B = 2.5 cases and state clearly whether the analytical formula is expected to apply there.
[Eq. (52)] The scheme is called 'Euler-Mayurama' but should be 'Euler-Maruyama'.
[Sec. 6.1.2] The word 'differes' should be 'differs'.
[References] References [11] and [12] are the same paper (Majda and Gershgorin, Phil. Trans. R. Soc. A, 2013); the duplication should be removed.
[Eq. (59) and Sec. 6] The set N in Eq. (59) is not defined; please state explicitly that the sum is over positive wavenumbers k = 1,2,... (or the finite set used in the simulations), and clarify whether the plotted analytical PDFs in Sec. 6 use eΣ(u) or the corrected 2eΣ(u).
[Sec. 5.1] The claim 'eΣ(u'_res) > 87 eΣ(u)' would be clearer if the comparison value of u (e.g., the mean zonal flow) were stated explicitly.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the tracer PDF is derived from the model equations and validated against independent simulations; self-citations are not load-bearing.

full rationale

The paper's central derivation is self-contained. The stationary conditional tracer variance eSigma_k(u) = alpha^2 E_v,k / (gamma_T,k^2 + omega_R,k(u)^2) is obtained in Proposition 4.2 and Appendix B.4 by taking the epsilon-to-0 limit of the exact trajectory conditional variance in Proposition 3.2. The mixture PDF in Theorem 4.3 is then a direct probabilistic consequence of conditional Gaussianity, and it is compared against independent numerical solutions of the SDE system. No target statistic is used as an input: parameter choices such as setting f = 0.4431 to place u'_res = -1 only define the experimental regime and do not enter the formula as fitted coefficients. The resonance condition omega_R,k = 0 is a derived peak of the Lorentzian variance, not an assumed definition of intermittency. The paper honestly reports that strong multiplicative noise (B = 2.5) degrades agreement with the analytical PDF, which further indicates that the analytical formula is not being hand-tuned to the data. Self-citations appear only in motivational or application-oriented passages and are not load-bearing for the derivation. A possible factor-of-two normalization issue in the printed sum over k in N would be a reproducibility or correctness concern, not a circularity, since it does not involve fitting a target quantity back into the derivation.

Assumptions & free parameters 10 free parameters · 7 assumptions · 0 invented entities

The central claim depends on a set of hand-chosen model parameters (timescale epsilon, zonal flow coefficients, shear dissipation, energy spectrum normalization) and on structural assumptions about periodicity, one-dimensional tracer dependence, white-noise forcing, the cubic zonal form, and the timescale separation. No new physical entities are introduced; the resonance condition is a consequence of the linear oscillator response, not an invented mechanism.

free parameters (10)

epsilon = 0.010
Timescale separation parameter chosen to make the multiscale approximation; controls the quasistatic limit in Prop 4.2.
f_linear = 0.4431
Zonal forcing in the OU model chosen so that the resonance threshold u'_res = -1 is crossed with probability 0.0228; a hand-set regime parameter.
gamma_u, sigma_u = 1, 1
Damping and noise amplitude of the linear zonal model; set to produce E_u = 0.5.
Nonlinear zonal coefficients (a, b, c, f, B, sigma_u) = varied (e.g., a=2 or -2, b=0, c=1, f=0 or 1, B=0 or 2.5, sigma_u=1)
Parameters of the cubic normal form with CAM noise; selected to produce bistable, skewed, or heavy-tailed zonal statistics in Figs. 1-2.
d_T = 0.1
Uniform tracer damping, added to control the zero mode; fixed for all experiments.
kappa = 0.001
Molecular diffusivity of the tracer; fixed for all experiments.
d_v, nu = 0.6, 0.1
Linear damping and viscosity of the shear flow; fixed for all experiments.
alpha = 1
Mean gradient amplitude; fixed to 1 as a normalization.
beta, F (QG) = 8.91, 2.5
Coriolis parameter and inverse squared deformation radius for the QG shear flow; chosen as representative geophysical values.
E0 = normalized
Energy spectrum amplitude; set by normalizing total energy equivalently between spectra, but absolute value not reported.

assumptions (7)

standard math Conditional Gaussian framework for linear dynamics given a fixed zonal flow path (Liptser-Shiryaev).
Invoked in Sec. 3 to derive the exact conditional trajectory solution and variance.
standard math Fubini's theorem and Ito isometry for stochastic integrals.
Used in the proofs of Props 3.1 and 3.2 in App. B.
domain assumption Velocity field is periodic in x, shear depends only on x, zonal flow is spatially uniform, and tracer fluctuations depend only on x.
Stated in Sec. 2 (Eq. (4)-(7)); this reduces the problem to a single spatial dimension and is the basis for the Fourier model.
domain assumption Shear flow is forced by white-in-time noise with a prescribed energy spectrum E_v,k.
Stated in Sec. 2.2 and Eq. (26); the Gaussianity of v depends on this linear stochastic forcing.
domain assumption Zonal flow dynamics follow the cubic normal form with CAM noise (Eq. (32)).
Taken from Majda et al. (2009) and used as the nonlinear zonal model in Sec. 2.3.2.
domain assumption Timescale separation: the velocity fields (u_t, v_t) evolve slowly compared to the tracer, encoded in the small parameter epsilon.
Introduced in Sec. 4.1 and used to derive the quasistatic stationary variance in Prop 4.2.
ad hoc to paper Explicit tracer damping d_T > 0 is added to compensate for the missing y-dependence.
Acknowledged in Sec. 2 (text near Eq. (7)); it is a modeling artifact needed to make the simplified model well-posed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Resonance-Driven Intermittency and Extreme Events in Turbulent Scalar Transport with a Mean Gradient." pith.science (2026). https://pith.science/paper/F3GI3S3C

@misc{pith2026250521688,
  author       = {Pith},
  title        = {Pith review of: Resonance-Driven Intermittency and Extreme Events in Turbulent Scalar Transport with a Mean Gradient},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F3GI3S3C}},
  note         = {Machine review of arXiv:2505.21688}
}

read the original abstract

We study the statistical properties of passive tracer transport in turbulent flows with a mean gradient, emphasizing tracer intermittency and extreme events. An analytically tractable model is developed, coupling zonal and shear velocity components with both linear and nonlinear stochastic dynamics. Formulating the model in Fourier space, a simple explicit solution for the tracer invariant statistics is derived. Through this model we identify the resonance condition responsible for non-Gaussian behavior and bursts in the tracer. Resonant conditions, that lead to a peak in the tracer variance, occur when the zonal flow and the shear flow phase speeds are equivalent. Numerical experiments across a range of regimes, including different energy spectra and zonal flow models, are performed to validate these findings and demonstrate how the velocity field and stochasticity determines tracer extremes. These results provide additional insight into the mechanisms underlying turbulent tracer transport, with implications for uncertainty quantification and data assimilation in geophysical and environmental applications.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

[12]

Elementary Models for Turbulent Diffusion with Complex Physical Features: EddyDiffusivity,SpectrumandIntermittency

A. J. Majda and B. Gershgorin. “Elementary Models for Turbulent Diffusion with Complex Physical Features: EddyDiffusivity,SpectrumandIntermittency”.PhilosophicalTransactionsoftheRoyalSocietyA:Mathematical, Physical and Engineering Sciences371.1982 (Jan. 13, 2013), p. 20120184

work page 1982
[1]

Elementary Models with Probability Distribution Function Intermittency for Passive Scalars with a Mean Gradient

A. Bourlioux and A. J. Majda. “Elementary Models with Probability Distribution Function Intermittency for Passive Scalars with a Mean Gradient”.Physics of Fluids14.2 (Feb. 1, 2002), pp. 881–897

work page 2002
[2]

Uncertainty Quantification of Nonlinear Lagrangian Data Assimilation Using Linear Stochastic Forecast Models

N. Chen and S. Fu. “Uncertainty Quantification of Nonlinear Lagrangian Data Assimilation Using Linear Stochastic Forecast Models”.Physica D: Nonlinear Phenomena452 (Oct. 2023), p. 133784

work page 2023
[3]

LagrangianDataAssimilationandParameterEstimationofanIdealizedSea IceDiscreteElementModel

N.Chen,S.Fu,andG.Manucharyan.“LagrangianDataAssimilationandParameterEstimationofanIdealizedSea IceDiscreteElementModel”.JournalofAdvancesinModelingEarthSystems13.10(Oct.2021),e2021MS002513

work page 2021
[4]

Filtering Nonlinear Turbulent Dynamical Systems through Conditional Gaussian Statistics

N. Chen and A. J. Majda. “Filtering Nonlinear Turbulent Dynamical Systems through Conditional Gaussian Statistics”.Monthly Weather Review144.12 (Dec. 1, 2016), pp. 4885–4917

work page 2016
[5]

Roots of polynomials of degrees 3 and 4

S. Janson. “Roots of Polynomials of Degrees 3 and 4”.arXiv:1009.2373(Sept. 13, 2010)

work page Pith review arXiv 2010
[6]

TheLocalStructureofTurbulenceinIncompressibleViscousFluidforVeryLargeReynolds

A.N.Kolmogorov.“TheLocalStructureofTurbulenceinIncompressibleViscousFluidforVeryLargeReynolds”. Numbers. In Dokl. Akad. Nauk SSSR30 (1941), p. 301

work page 1941
[7]

Stochastic Superparameterization and Multiscale Filtering of Turbulent Tracers

Y. Lee, A. J. Majda, and D. Qi. “Stochastic Superparameterization and Multiscale Filtering of Turbulent Tracers”. Multiscale Modeling & Simulation15.1 (Jan. 2017), pp. 215–234

work page 2017

Show all 23 references

[8]

R. S. Liptser and A. N. Shiryaev.Statistics of Random Processes II. Applications. 2nd ed. Stochastic Modelling and Applied Probability 6. Springer-Verlag Berlin Heidelberg, 2001

2001
[9]

Intermittency in Turbulent Diffusion Models with a Mean Gradient

A. J. Majda and X. T. Tong. “Intermittency in Turbulent Diffusion Models with a Mean Gradient”.Nonlinearity 28.11 (Oct. 1, 2015), pp. 4171–4208. 18

2015
[10]

NormalFormsforReducedStochasticClimateModels

A.J.Majda,C.Franzke,andD.Crommelin.“NormalFormsforReducedStochasticClimateModels”.Proceedings of the National Academy of Sciences106.10 (Mar. 10, 2009), pp. 3649–3653

2009
[13]

Simplified Models for Turbulent Diffusion: Theory, Numerical Modelling, and Physical Phenomena

A. J. Majda and P. R. Kramer. “Simplified Models for Turbulent Diffusion: Theory, Numerical Modelling, and Physical Phenomena”.Physics Reports314.4–5 (June 1999), pp. 237–574

1999
[14]

SimpleExampleswithFeaturesofRenormalizationforTurbulentTransport

MarcoAvellanedaandA.J.Majda.“SimpleExampleswithFeaturesofRenormalizationforTurbulentTransport”. PhilosophicalTransactionsoftheRoyalSocietyofLondon.SeriesA:PhysicalandEngineeringSciences346.1679 (Feb. 15, 1994), pp. 205–233

1994
[15]

Recovering the Eulerian Energy Spectrum from Noisy Lagrangian Tracers

M. A. Mohamad and A. J. Majda. “Recovering the Eulerian Energy Spectrum from Noisy Lagrangian Tracers”. Physica D: Nonlinear Phenomena403 (Feb. 1, 2020), p. 132374

2020
[16]

Probabilistic Description of Extreme Events in Intermittently Unstable DynamicalSystemsExcitedbyCorrelatedStochasticProcesses

M. A. Mohamad and T. P. Sapsis. “Probabilistic Description of Extreme Events in Intermittently Unstable DynamicalSystemsExcitedbyCorrelatedStochasticProcesses”.SIAM/ASAJournalonUncertaintyQuantification 3.1 (Jan. 2015), pp. 709–736

2015
[17]

Long Tails in Deep Columns of Natural and Anthropogenic Tropospheric Tracers

J. D. Neelin, B. R. Lintner, B. Tian, Q. Li, L. Zhang, P. K. Patra, M. T. Chahine, and S. N. Stechmann. “Long Tails in Deep Columns of Natural and Anthropogenic Tropospheric Tracers”.Geophysical Research Letters37.5 (Mar. 1, 2010)

2010
[18]

Atmospheric Diffusion Shown on a Distance-Neighbour Graph

L. F. Richardson. “Atmospheric Diffusion Shown on a Distance-Neighbour Graph”.Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character110.756 (Apr. 1926), pp. 709–737

1926
[19]

Tracer Transport along and across Coherent Jets in Two-Dimensional Turbulent Flow

K. S. Smith. “Tracer Transport along and across Coherent Jets in Two-Dimensional Turbulent Flow”.Journal of Fluid Mechanics544 (Dec. 2005), pp. 133–142

2005
[20]

Turbulent Mixing: A Perspective

K. R. Sreenivasan. “Turbulent Mixing: A Perspective”.Proceedings of the National Academy of Sciences116.37 (Sept. 10, 2019), pp. 18175–18183

2019
[21]

Diffusion by Continuous Movements

G. I. Taylor. “Diffusion by Continuous Movements”.Proceedings of the London Mathematical Societys2-20.1 (1922), pp. 196–212

1922
[22]

G. K. Vallis.Atmospheric and Oceanic Fluid Dynamics: Fundamentals and Large-Scale Circulation. Cambridge: Cambridge university press, 2006

2006
[23]

PassiveScalarsinTurbulentFlows

Z.Warhaft.“PassiveScalarsinTurbulentFlows”.AnnualReviewofFluidMechanics32(Volume32,2000Jan.1, 2000), pp. 203–240

2000
[24]

Invariant Measures and Asymptotic Gaussian Bounds for Normal Forms of Stochastic Climate Model

Y. Yuan and A. J. Majda. “Invariant Measures and Asymptotic Gaussian Bounds for Normal Forms of Stochastic Climate Model”.Chinese Annals of Mathematics, Series B32.3 (May 2011), pp. 343–368. 19 A Zonal Flow Model Details A.1 Dynamical regimes To study the dynamical regimes of ...

2011

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [12]

Elementary Models for Turbulent Diffusion with Complex Physical Features: EddyDiffusivity,SpectrumandIntermittency

A. J. Majda and B. Gershgorin. “Elementary Models for Turbulent Diffusion with Complex Physical Features: EddyDiffusivity,SpectrumandIntermittency”.PhilosophicalTransactionsoftheRoyalSocietyA:Mathematical, Physical and Engineering Sciences371.1982 (Jan. 13, 2013), p. 20120184

work page 1982

[2] [1]

Elementary Models with Probability Distribution Function Intermittency for Passive Scalars with a Mean Gradient

A. Bourlioux and A. J. Majda. “Elementary Models with Probability Distribution Function Intermittency for Passive Scalars with a Mean Gradient”.Physics of Fluids14.2 (Feb. 1, 2002), pp. 881–897

work page 2002

[3] [2]

Uncertainty Quantification of Nonlinear Lagrangian Data Assimilation Using Linear Stochastic Forecast Models

N. Chen and S. Fu. “Uncertainty Quantification of Nonlinear Lagrangian Data Assimilation Using Linear Stochastic Forecast Models”.Physica D: Nonlinear Phenomena452 (Oct. 2023), p. 133784

work page 2023

[4] [3]

LagrangianDataAssimilationandParameterEstimationofanIdealizedSea IceDiscreteElementModel

N.Chen,S.Fu,andG.Manucharyan.“LagrangianDataAssimilationandParameterEstimationofanIdealizedSea IceDiscreteElementModel”.JournalofAdvancesinModelingEarthSystems13.10(Oct.2021),e2021MS002513

work page 2021

[5] [4]

Filtering Nonlinear Turbulent Dynamical Systems through Conditional Gaussian Statistics

N. Chen and A. J. Majda. “Filtering Nonlinear Turbulent Dynamical Systems through Conditional Gaussian Statistics”.Monthly Weather Review144.12 (Dec. 1, 2016), pp. 4885–4917

work page 2016

[6] [5]

Roots of polynomials of degrees 3 and 4

S. Janson. “Roots of Polynomials of Degrees 3 and 4”.arXiv:1009.2373(Sept. 13, 2010)

work page Pith review arXiv 2010

[7] [6]

TheLocalStructureofTurbulenceinIncompressibleViscousFluidforVeryLargeReynolds

A.N.Kolmogorov.“TheLocalStructureofTurbulenceinIncompressibleViscousFluidforVeryLargeReynolds”. Numbers. In Dokl. Akad. Nauk SSSR30 (1941), p. 301

work page 1941

[8] [7]

Stochastic Superparameterization and Multiscale Filtering of Turbulent Tracers

Y. Lee, A. J. Majda, and D. Qi. “Stochastic Superparameterization and Multiscale Filtering of Turbulent Tracers”. Multiscale Modeling & Simulation15.1 (Jan. 2017), pp. 215–234

work page 2017