pith. sign in

arxiv: 1907.11553 · v1 · pith:3ZLBJIJEnew · submitted 2019-07-25 · 🧮 math.PR

Spatial ergodicity for SPDEs via Poincar\'e-type inequalities

Le Chen , Davar Khoshnevisan , David Nualart , Fei Pu This is my paper

Pith reviewed 2026-05-24 16:07 UTC · model grok-4.3

classification 🧮 math.PR
keywords spatial ergodicitystochastic PDEsPoincaré inequalitiesintermittencyGaussian noiseMalliavin calculusstationary processesmixing
0
0 comments X

The pith

For parabolic SPDEs started at the constant one, the spatial field x to u(t,x) is stationary and ergodic at every t>0 when the noise correlation decays mildly.

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

The paper proves that solutions to the parabolic SPDE partial_t u = 1/2 Delta u + sigma(u) eta, started from u(0) identically one, yield a spatially stationary and ergodic random field at each fixed positive time whenever the spatial correlation f of the Gaussian noise satisfies a mild decay condition at infinity. A reader would care because the result supplies the missing rigorous link between spatial ergodicity and the intermittent structure of solutions that had been discussed heuristically in the literature. The argument rests on new Poincaré-type inequalities derived via harmonic analysis of positive definite functions together with Malliavin calculus; these inequalities also deliver mixing of the field and a proof of a conjecture on the spatial size of intermittency islands. The conclusions are sharp in that they recover the classical ergodicity theorem for Gaussian fields precisely when the nonlinearity sigma is constant.

Core claim

If the initial condition is identically one and the spatial correlation function f of the noise satisfies a mild decay condition, then the solution u(t,x) is a stationary and ergodic random field in the spatial variable x for every fixed t>0. The same set of Poincaré inequalities also ensures that u(t) is mixing for every t>0 and settles a conjecture of Conus et al. on the diameter of intermittency islands.

What carries the argument

Poincaré-type inequalities for the random field u(t,·) obtained from harmonic analysis of functions of positive type and Malliavin calculus, which bound the variance of spatial averages and thereby yield ergodicity.

If this is right

  • The random field u(t) is mixing for every t>0 under the stated conditions.
  • A conjecture of Conus et al. about the size of intermittency islands is proved.
  • The results recover the classical ergodicity theory of Maruyama when sigma is constant.
  • These facts give rigorous justification for heuristic arguments connecting spatial ergodicity to intermittency.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The decay condition on f is likely sharp, since its violation would allow persistent long-range correlations that prevent ergodicity.
  • The same Poincaré inequalities may apply directly to other SPDEs with colored noise to obtain spatial mixing or ergodicity.
  • Stationarity plus ergodicity together imply that spatial averages of u(t,·) converge almost surely to a deterministic constant as the domain expands.

Load-bearing premise

The spatial correlation function f of the noise must satisfy a mild decay condition at large distances.

What would settle it

A correlation function f violating the mild decay condition for which the solution starting from the constant one fails to be ergodic at some positive time would falsify the claim.

read the original abstract

Consider a parabolic stochastic PDE of the form $\partial_t u=\frac{1}{2}\Delta u + \sigma(u)\eta$, where $u=u(t\,,x)$ for $t\ge0$ and $x\in\mathbb{R}^d$, $\sigma:\mathbb{R}\rightarrow\mathbb{R}$ is Lipschitz continuous and non random, and $\eta$ is a centered Gaussian noise that is white in time and colored in space, with a possibly-signed homogeneous spatial correlation $f$. If, in addition, $u(0)\equiv1$, then we prove that, under a mild decay condition on $f$, the process $x\mapsto u(t\,,x)$ is stationary and ergodic at all times $t>0$. It has been argued that, when coupled with moment estimates, spatial ergodicity of $u$ teaches us about the intermittent nature of the solution to such SPDEs \cite{BertiniCancrini1995,KhCBMS}. Our results provide rigorous justification of such discussions. Our methods hinge on novel facts from harmonic analysis and functions of positive type, as well as from Malliavin calculus and Poincar\'e inequalities. We further showcase the utility of these Poincar\'e inequalities by: (a) describing conditions that ensure that the random field $u(t)$ is mixing for every $t>0$; and by (b) giving a quick proof of a conjecture of Conus et al \cite{CJK12} about the "size" of the intermittency islands of $u$. The ergodicity and the mixing results of this paper are sharp, as they include the classical theory of Maruyama \cite{Maruyama} (see also Dym and McKean \cite{DymMcKean}) in the simple setting where the nonlinear term $\sigma$ is a constant function.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The paper proves that solutions to the parabolic SPDE ∂_t u = (1/2) Δu + σ(u) η with initial condition u(0) ≡ 1 are spatially stationary and ergodic for every t > 0, provided the spatial correlation f of the Gaussian noise satisfies a mild decay condition. The argument proceeds by establishing novel Poincaré-type inequalities via Malliavin calculus and harmonic analysis on functions of positive type; these yield the spectral gap needed for ergodicity. The same inequalities are used to prove mixing of the field u(t) for each t > 0 and to give a short proof of the Conus–Joseph–Khoshnevisan conjecture on the size of intermittency islands. The results are shown to be sharp by recovering the classical Maruyama ergodicity theorem in the linear case σ ≡ const.

Significance. If the derivations hold, the work supplies the first rigorous spatial-ergodicity statement for genuinely nonlinear multiplicative noise with colored spatial correlation, thereby justifying heuristic discussions of intermittency that rely on ergodicity. The new Poincaré inequalities appear to be of independent interest and may apply to other SPDEs or Gaussian multiplicative chaos models. Explicit reduction to the Maruyama–Dym–McKean theorem when σ is constant provides a valuable internal consistency check and demonstrates that the decay hypothesis on f is essentially optimal.

minor comments (3)
  1. [Abstract] Abstract, line 4: the phrase 'mild decay condition on f' is used without a quantitative statement; a one-sentence description of the precise integrability or Fourier-decay requirement would improve readability for readers who do not reach the main theorem statement.
  2. The reduction to the constant-σ case is invoked as a sanity check, but the manuscript does not explicitly identify the section or equation where the classical Maruyama argument is recovered; adding a short dedicated paragraph or remark would strengthen the claim of sharpness.
  3. Notation: the spatial correlation is denoted f throughout, yet the abstract and introduction occasionally refer to it as the 'spatial correlation function f of the noise'; a single consistent phrase would eliminate minor ambiguity.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, significance assessment, and recommendation of minor revision. The report correctly identifies the main contributions, including the novel Poincaré inequalities, the ergodicity result, the mixing property, the short proof of the Conus–Joseph–Khoshnevisan conjecture, and the consistency check with Maruyama’s theorem. No major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper establishes its central claims of spatial stationarity and ergodicity for the SPDE solution at all t>0 by deriving novel Poincaré-type inequalities via Malliavin calculus and harmonic analysis applied to functions of positive type, under an explicit mild decay hypothesis on the spatial correlation f. These inequalities are not obtained by fitting or self-definition but are proved directly as new facts. The reduction to the classical Maruyama ergodicity result when σ is constant functions as an external consistency check rather than a load-bearing input. No self-citations are used to justify uniqueness or core premises, no ansatzes are smuggled, and no known results are merely renamed; the derivation chain remains independent and self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The paper relies on standard domain assumptions from SPDE theory; no free parameters or invented entities appear in the abstract.

axioms (3)
  • domain assumption Existence and uniqueness of mild solutions to the parabolic SPDE with Lipschitz coefficient σ
    Standard assumption invoked for the well-posedness of the equation under consideration.
  • domain assumption The driving noise η is a centered Gaussian random measure that is white in time and colored in space with homogeneous correlation f
    Given in the problem formulation and used throughout the harmonic analysis arguments.
  • domain assumption The initial condition u(0,x) ≡ 1 for all x
    Explicitly stated as the setting in which stationarity and ergodicity are proved.

pith-pipeline@v0.9.0 · 5885 in / 1464 out tokens · 40205 ms · 2026-05-24T16:07:01.012916+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

56 extracted references · 56 canonical work pages · 3 internal anchors

  1. [1]

    The sto chastic heat equation: Feynman-Kac formula and intermittence

    Bertini, Lorenzo and Nicoletta Cancrini (1995). The sto chastic heat equation: Feynman-Kac formula and intermittence. J. Statist. Phys. 78(5-6) 1377–1401

    work page 1995

  2. [2]

    Burkholder, D. L. (1966). Martingale transforms. Ann. Math. Statist. 37 1494–1504

    work page 1966

  3. [3]

    L., Davis, B

    Burkholder, D. L., Davis, B. J., and Gundy, R. F. (1972). I ntegral inequalities for convex functions of operators on martingales. In: Proceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability II, 223–240, University of California Press, Berkeley, California

    work page 1972

  4. [4]

    Burkholder, D. L. and Gundy, R. F. (1970). Extrapolation and interpolation of quasi-linear operators on martingales. Acta Math. 124 249–304

    work page 1970

  5. [5]

    Hsu, and M

    Capitaine, Mirielle, Elton P. Hsu, and M. Ledoux (1997). Martingale representation and a simple proof of the logarithmic Sobolev inequalities on pat h space. Electr. Comm. Probab. 2 71–81

    work page 1997

  6. [6]

    Cardon-Weber, C. and A. Millet (2004). On strongly Petro vski ˘ ı’s parabolic SPDEs in arbi- trary dimension and application to the stochastic Cahn-Hil liard equation. J. Theoret. Probab. 17(1) 1–49

    work page 2004

  7. [7]

    Carlen, Eric and Paul Kr´ ee (1991).Lp estimates on iterated stochastic integrals. Ann. Probab. 19(1) 354–368

    work page 1991

  8. [8]

    Carmona, Ren´ e A. and S. A. Molchanov (1994). Parabolic A nderson Problem and Inter- mittency, Memoires of the Amer. Math. Soc. 108, American Mathematical Society, Rhode Island

    work page 1994

  9. [9]

    Two-poi nt correlation function and Feynman-Kac formula for the stochastic heat equation

    Chen, Le, Yaozhong Hu, and David Nualart (2017). Two-poi nt correlation function and Feynman-Kac formula for the stochastic heat equation. Potential Anal. 46(4) 779–797. 37

    work page 2017

  10. [10]

    Comparison principle for stochastic heat equation on Rd

    Chen, Le and Jingyu Huang (2019). Comparison principle for stochastic heat equation on Rd. Ann. Probab. 47(2) 989-1035

    work page 2019

  11. [11]

    Regularity and strict positivity of densities for the stochastic heat equation on $\mathbb{R}^d$

    Chen, Le and Jingyu Huang (2019). Regularity and strict positivity of densities for the stochastic heat equation on Rd. Preprint available at https://arxiv.org/abs/1902.02382

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  12. [12]

    Spatial asymptotics for the paraboli c Anderson models with generalized time-space Gaussian noise

    Chen, Xia (2016). Spatial asymptotics for the paraboli c Anderson models with generalized time-space Gaussian noise. Ann. Probab. 44(2) 1535–1598

    work page 2016

  13. [13]

    Spati al asymptotics for the parabolic Anderson model driven by a Gaussian rough noise

    Chen, Xia, Yaozhong Hu, and David Nualart (2017). Spati al asymptotics for the parabolic Anderson model driven by a Gaussian rough noise. Electron. J. Probab. 22(65) 38 pp

    work page 2017

  14. [14]

    Malliavin calculus and stochastic i ntegral representation of diffusion processes

    Ocone, Dan (1984). Malliavin calculus and stochastic i ntegral representation of diffusion processes. Stochastics 12 161–185

    work page 1984

  15. [15]

    Correlation-length bounds, and estimates for intermittent islands in parabolic SPDEs

    Conus, Daniel, Mathew Joseph, and Davar Khoshnevisan ( 2012). Correlation-length bounds, and estimates for intermittent islands in parabolic SPDEs. Electr. J. Probab. 17(102) 1–15

    work page 2012

  16. [16]

    On the exi stence and position of the farthest peaks of a family of stochastic heat and wave equations

    Conus, Daniel and Davar Khoshnevisan (2012). On the exi stence and position of the farthest peaks of a family of stochastic heat and wave equations. Probab. Theory Rel. Fields 152(3-4) 681–701

    work page 2012

  17. [17]

    On the chaotic character of the stochastic heat equation, before the onset of intermi tttency

    Conus, Daniel, Mathew Joseph, and Davar Khoshnevisan, (2013). On the chaotic character of the stochastic heat equation, before the onset of intermi tttency. Ann. Probab. 41(3B) 2225–2260

    work page 2013

  18. [18]

    On the chaotic character of the stochastic heat equation, II

    Conus, Daniel, Mathew Joseph, Davar Khoshnevisan, and Shang-Yuan Shiu (2013). On the chaotic character of the stochastic heat equation, II. Probab. Theory Related Fields 156(3-4) 483–533

    work page 2013

  19. [19]

    Dalang, Robert C. (1999). Extending the martingale mea sure stochastic integral with appli- cations to spatially homogeneous s.p.d.e.’s. Electron. J. Probab. 4(6), 29 pp

    work page 1999

  20. [20]

    Dalang, Robert C. and N. E. Frangos (1998). The stochast ic wave equation in two dimensions. Ann. Probab. 26(1) 187–212

    work page 1998

  21. [21]

    On the Lp norms of stochastic integrals and other martingales

    Davis, Burgess (1976). On the Lp norms of stochastic integrals and other martingales. Duke Math. J. 43(4) 697–704

    work page 1976

  22. [22]

    Doob, J. L. (1990). Stochastic Processes. Reprint of the 1953 original. John Wiley & Sons, Inc., New York, viii+654

    work page 1990

  23. [23]

    Dym, H. and H. P. McKean (1976). Gaussian Processes, Function Theory, and the Inverse Spectral Problem. Academic Press. New York

    work page 1976

  24. [24]

    A., and Louis Sucheston (1992)

    Edgar, G. A., and Louis Sucheston (1992). Stopping Times and Directed Processes. Cambridge University Press, Cambridge

    work page 1992

  25. [25]

    Inte rmittence and nonlinear parabolic stochastic partial differential equations

    Foondun, Mohammud and Davar Khoshnevisan (2009). Inte rmittence and nonlinear parabolic stochastic partial differential equations. Electron. J. Probab. 14(21) 548–568

    work page 2009

  26. [26]

    On th e stochastic heat equation with spatially-colored random forcing

    Foondun, Mohammud and Davar Khoshnevisan (2013). On th e stochastic heat equation with spatially-colored random forcing. Trans. Amer. Math. Soc. 365(1) 409–458. 38

    work page 2013

  27. [27]

    L’int´ egr ale stochastique comme op´ erateur de divergence dans l’espace fonctionnel

    Gaveau, Bernard and Trauber, Philip (1982). L’int´ egr ale stochastique comme op´ erateur de divergence dans l’espace fonctionnel. J. Functional Anal. 46 230–238

    work page 1982

  28. [28]

    Potential theory of L´ evy process es

    Hawkes, John (1979). Potential theory of L´ evy process es. Proc. London Math. Soc. (3) 38, no. 2, 335–352

    work page 1979

  29. [29]

    Some geometric aspects of potenti al theory

    Hawkes, John (1984). Some geometric aspects of potenti al theory. In: Stochastic Analysis and Applications (Swansea, 1983), 130–154, Lecture Notes in Math. 1095 Springer, Berlin

    work page 1984

  30. [30]

    Stochastic heat equation with rough dependence in space

    Hu, Yaozhong, Jingyu, Huang, Khoa Lˆ e, David Nualart, a nd Samy Tindel (2017). Stochastic heat equation with rough dependence in space. Ann. Probab. 45(6B) 4561–4616

    work page 2017

  31. [31]

    Stochastic heat equa- tions with general multiplicative Gaussian noises: H¨ older continuity and intermittency

    Hu, Yaozhong, Jingyu Huang, David Nualart, and Samy Tin del (2015). Stochastic heat equa- tions with general multiplicative Gaussian noises: H¨ older continuity and intermittency. Elec- tron. J. Probab. 20(55) 50 pp

    work page 2015

  32. [32]

    Large time asymptotics for the parabolic Anderson model driven by space and time correlated noise

    Huang, Jingyu, Khoa Lˆ e, and David Nualart, (2017). Large time asymptotics for the parabolic Anderson model driven by space and time correlated noise. Stoch. Partial Differ. Equ. Anal. Comput. 5(4) 614–651

    work page 2017

  33. [33]

    A Central Limit Theorem for the stochastic heat equation

    Huang, Jingyu, David Nualart, and Lauri Viitasaari (20 18). A central limit theorem for the stochastic heat equation. Preprint available at https://arxiv.org/abs/1810.09492

    work page internal anchor Pith review Pith/arXiv arXiv

  34. [34]

    Gaussian fluctuations for the stochastic heat equation with colored noise

    Huang, Jingyu, David Nualart, Lauri Viitasaari, and Gu angqu Zheng (2019). Gaussian fluctuations for the stochastic heat equation with colored n oise. Preprint available at https://arxiv.org/abs/1903.02509

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  35. [35]

    Some Random Series of Functions

    Kahane, Jean-Pierre (1985). Some Random Series of Functions. Second edition. Cambridge University Press, Cambridge

    work page 1985

  36. [36]

    A note on sto chastic wave equations

    Karczewska, Anna and Jerzy Zabczyk (2001). A note on sto chastic wave equations. In: Evo- lution Equations and Their Applications in Physical and Life S ciences (Bad Herrenalb, 1998), 501–511, Lecture Notes in Pure and Appl. Math. 215 Dekker, NY

    work page 2001

  37. [37]

    Analysis of Stochastic Partial Differential Equations

    Khoshnevisan, Davar (2014). Analysis of Stochastic Partial Differential Equations. CBMS Regional Conference Series in Mathematics, 119. American M athematical Society, Providence, RI, 2014. viii+116

    work page 2014

  38. [38]

    Multiparameter Processes

    Khoshnevisan, Davar (2002). Multiparameter Processes. Springer-Verlag, New York, xx+584

    work page 2002

  39. [39]

    Intermittency and multifractal- ity: A case study via stochastic PDEs

    Kim, Kunwoo, Davar Khoshnevisan, and Yimin Xiao (2017) . Intermittency and multifractal- ity: A case study via stochastic PDEs. Ann. Probab. 45(6A) 3697–3751

    work page 2017

  40. [40]

    A macr oscopic multifractal analysis of parabolic stochastic PDEs

    Kim, Kunwoo, Davar Khoshnevisan, and Yimin Xiao. A macr oscopic multifractal analysis of parabolic stochastic PDEs. Comm. Math. Phys. 360 (2018) 307–346

    work page 2018

  41. [41]

    L´ epingle, Dominique and Ouvrard, J. Y. (1973). Martin gales browniennes hilbertiennes. C. R. Acad. Sci. Paris S´ er. A-B 276 A1225–A1228

    work page 1973

  42. [42]

    The harmonic analysis of sta tionary stochastic processes

    Maruyama, Gisir¯ o (1949). The harmonic analysis of sta tionary stochastic processes. Mem. Faculty Sci. Kyushu Univ. Ser. A. 4 45–106

    work page 1949

  43. [43]

    A stochastic wave equation in two space dimension: smoothness of the law

    Millet, Annie and Marta Sanz-Sol´ e (1999). A stochastic wave equation in two space dimension: smoothness of the law. Ann. Probab. 27(2) 803–844. 39

    work page 1999

  44. [44]

    On the support of solutions to the heat equation with noise

    Mueller, Carl (1991). On the support of solutions to the heat equation with noise. Stoch. & Stoch. Rep. 37(4) 225–245

    work page 1991

  45. [45]

    Some tools and results for parabo lic stochastic partial differential Equa- tions

    Mueller, Carl (2009). Some tools and results for parabo lic stochastic partial differential Equa- tions. In: A Minicourse on Stochastic Partial Differential Equations , pp. 111–144, Lecture Notes in Math. 1962 Springer, Berlin

    work page 2009

  46. [46]

    The Malliavin Calculus and Related Topics

    Nualart, David (2006). The Malliavin Calculus and Related Topics . Springer

    work page 2006

  47. [47]

    and Clark, C.W

    Olver, F.W.J., Lozier, D.W., Boisvert, R.F. and Clark, C.W. (2010). NIST Handbook of Mathematical Functions. Cambridge Univ. Press, Cambridge, UK

    work page 2010

  48. [48]

    Ouvrard, J. Y. (1975/76). Repr´ esentation de martinga les vectorielles de carr´ e int´ egrable ` a valeurs dans des espaces de Hilbert r´ eels s´ eparables.Z. Wahrscheinlichkeitstheorie und Verw. Gebiete 33(3) 195–208

    work page 1975

  49. [49]

    The Cauchy problem for a nonline ar stochastic wave equation in any dimension

    Peszat, Szymon (2002). The Cauchy problem for a nonline ar stochastic wave equation in any dimension. J. Evol. Equ. 2(3) 383–394

    work page 2002

  50. [50]

    Nonlinear sto chastic wave and heat equations

    Peszat, Szymon and Jerzy Zabczyk (2000). Nonlinear sto chastic wave and heat equations. Probab. Theory Related Fields 116(3) 421–443

    work page 2000

  51. [51]

    Ergodic Theory

    Peterson, Karl (1990). Ergodic Theory. Cambridge University Press, Cambridge

    work page 1990

  52. [52]

    Stein, Elias M. (1970). Singular Integrals and Differentiability Properties of Functions. Prince- ton University Press, Princeton, N.J

    work page 1970

  53. [53]

    Walsh, John B. (1986). An Introduction to Stochastic Partial Differential Equation s. `Ecole d’´ et´ e de probabilit´ es de Saint-Flour, XIV-1984, 265–439, In: Lecture Notes in Math. 1180, Springer, Berlin

    work page 1986

  54. [54]

    Zeldovich, Ya. B., S. A. Molchanov, A. A. Ruzmaikin, and D. D. Sokolov (1988). Intermit- tency, diffusion, and generation in a nonstationary random me dium, Sov. Sci. Rev. C. Math. Phys. 7 1–110

    work page 1988

  55. [55]

    Zeldovich, Ya. B., S. A. Molchanov, Ruzmaikin, A. A., an d Sokolov, D. D. (1985). Intermit- tency of passive fields in random media, J. of Experimental and Theoretical Physics [actual journal title: Жurnal eksperimentalьno/ishort i teoretiqesko/ishort fiziki] 89[6(12)] 2061–2072. (In Russian)

    work page 1985

  56. [56]

    B., Ruzmaikin, A

    Zeldovich, Ya. B., Ruzmaikin, A. A., and Sokoloff, D. D. (1 990). The Almighty Chance, World Scientific Lecture Notes in Physics, vol. 20, Singapore,. 40

    work page