Pith. sign in

REVIEW 2 major objections 7 minor 38 references

Semiclassical Schr\"odinger operators with purely imaginary potential

T0 review · 2 major / 7 minor · reviewed 2026-07-09 · glm-5.2

Pith's one-line read Degenerate critical points control spectra of non-self-adjoint Schrödinger operators

desk verdict Solid extension of non-selfadjoint semiclassical spectral asymptotics to degenerate critical points; the main theorem is conditionally correct but rests on an incompletely verified model operator. read the letter →

arxiv 2607.07301 v1 pith:NFPMONPB submitted 2026-07-08 math.AP math-phmath.MPmath.SP

classification math.APmath-phmath.MPmath.SP
keywords potentialassociatedcriticalimaginaryodingeroperatorspointspurely
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

For the operator P = -h²Δ + iV(x) on a bounded domain, where V is real-valued and the semiclassical parameter h → 0, the eigenvalues closest to the imaginary axis are determined not by the global shape of V but by the local structure of V at its most degenerate critical points. When V near a critical point c is approximated by a homogeneous polynomial V_c of degree α_c, the leftmost eigenvalues near the energy iE take the form λ = iE + h^σ μ + o(h^σ), where σ = 2α/(α+2), α is the maximal vanishing order among critical points at energy E, and μ ranges over the eigenvalues of the model operator P_c = -Δ + iV_c. The resolvent is bounded by C h^{-σ} away from these quasi-eigenvalues, which is optimal. The result extends from the standard self-adjoint theory (where the max-min principle and quadratic approximations suffice) to the non-self-adjoint setting, where the resolvent is not controlled by distance to the spectrum and pseudospectral effects dominate.

What carries the argument

The proof rests on three pillars: (1) a scaling conjugation U_c that relates the semiclassical operator P near a critical point c to the h-independent model operator P_c, extracting the scaling exponent σ = 2α/(α+2); (2) an auxiliary operator Q = P + G², where G is a cutoff localizing near maximally degenerate critical points, which is elliptic where P is not and satisfies the resolvent bound ||(Q-z)^{-1}|| ≤ C h^{-σ}; (3) a parametrix R(z) = Σ φ_c (P_c - z)^{-1} φ_c + ψ (Q-z)^{-1} ψ that patches local model resolvents with the global Q-resolvent via a partition of unity. The rank of the spectral projector of P near each quasi-eigenvalue is shown to equal the sum of ranks of the model Spectr

What would settle it

A potential V satisfying the approximation condition (1.6) but violating the non-degeneracy condition (1.5), such as V(x₁,x₂) = x₁² in R², where the model operator -Δ + ix₁² has non-discrete spectrum and the eigenvalue asymptotic of Theorem 1.5 cannot hold.

Watch

Extended reading notes

Core claim

The central mechanism is a scaling argument combined with a parametrix construction. Near each maximally degenerate critical point c, the operator P is conjugated by a scaling isometry U_c that zooms in at scale h^β (where β = 2/(α+2)). Under this scaling, P decomposes as h^σ P_c + iE plus lower-order terms, where P_c = -Δ + iV_c is a scale-invariant model operator. The full operator's spectrum near iE is then captured by patching together local resolvents of the model operators P_c at each critical point with a global resolvent estimate for an auxiliary operator Q (which is P plus a confining term G² that makes it elliptic near critical points). The key resolvent estimate for Q uses a Poinc

Load-bearing premise

The non-degeneracy condition requiring that the homogeneous leading part V_c of the potential has no critical points away from c itself (i.e., ∇V_c(x) ≠ 0 for x ≠ c). This ensures critical points are isolated and the level-set geometry is controllable, but it excludes physically relevant potentials like V = x₁² in two dimensions, where the model operator has non-discrete spectrum and the main theorem fails.

Editorial extensions

If this is right

  • The decay rate of solutions to the heat equation ∂_t u + Pu = 0 is governed by the spectral gap μ₀ = inf Re(Λ), where Λ is the union of model operator spectra, giving the sharp exponential decay rate h^σ μ₀ for the semigroup e^{-tP}.
  • For shear flows in fluid mechanics governed by ∂_t u + V(x)∂_y u - νΔ u = 0, the enhanced dissipation rate is ν^{α/(α+2)} |k|^{2α/(α+2)}, with the precise prefactor determined by the model operator spectrum.
  • The framework extends to general complex-valued potentials V with min Re V = 0, where the leftmost eigenvalues should be determined by points where Re V = 0 and Im V' = 0 simultaneously.
  • The C¹ regularity requirement (rather than C∞) allows treatment of degeneracies of arbitrary order α > 1, yielding the full range of scaling exponents σ ∈ (2/3, 2) for the eigenvalue asymptotics.
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

2 major / 7 minor

Summary. The paper studies the semiclassical Schrödinger operator $P = -h^2 Delta + iV(x)$ on a bounded domain (or torus), where $V$ is a real-valued $C^1$ potential. Under a 'degenerate Morse' assumption (Assumption 1) — requiring that near each critical point $c$ of energy $E$, the potential is approximated by a homogeneous polynomial $V_c$ of degree $alpha_c > 1$ satisfying a non-degeneracy condition on $nabla V_c$ — the authors prove that the leftmost eigenvalues of $P$ near $iE$ are asymptotically given by $iE + h^sigma mu_j + o(h^sigma)$, where $sigma = 2alpha/(alpha+2)$, $alpha$ is the maximal vanishing order at energy $E$, and $mu_j$ are eigenvalues of the model operators $P_c = -Delta + iV_c$ associated to the most degenerate critical points (Theorem 1.5). A resolvent estimate $|(P-z)^{-1}| leq C h^{-sigma}$ away from quasi-eigenvalues is also established. Applications to the associated heat equation and shear flow dissipation are given (Theorem 1.17). The proof proceeds via a parametrix construction combining local model resolvents with an auxiliary elliptic operator $Q = P + G^2$, following the strategy of Coti Zelati–Gallay [16] and Henry [26].

Significance. The paper makes a solid contribution to the spectral theory of non-selfadjoint semiclassical operators. The degenerate Morse setting goes beyond the standard Morse (quadratic) case and allows for $C^1$ potentials with critical points of arbitrary finite order, yielding a continuum of scaling exponents $sigma in (2/3, 2)$. The identification of the optimal spectral gap constant as the infimum of the real part of the model operator spectra, and the eigenmode expansion for the semigroup, are concrete improvements over prior results of [16]. The connection to enhanced dissipation in shear flows provides a natural physical motivation. The authors are transparent about the limitations of the model operator theory in the general (non-signed) case, which is appropriate.

major comments (2)
  1. Proposition 1.15(ii) asserts a polynomial resolvent bound $|(mathcal{P}-z)^{-1}| leq C + C/text{dist}(z, sigma(mathcal{P}))^N$ for the model operator $mathcal{P} = -Delta + iV$ in the general case (homogeneous $V$ satisfying (1.5), not necessarily signed). The proof given in Section 3.3 only covers part (i) (the large-$|z|$ regime) in general; for part (ii), the argument at the end of Section 3.3 invokes the discreteness of the spectrum (Proposition 1.12) to obtain (3.28), but the polynomial blow-up rate $N$ is not actually established — it is merely asserted to exist. This estimate is used implicitly in the spectral projector construction (via (2.14)–(2.16) and the contour integrals defining $Pi$ and $Pi_c$). While the rank argument in Section 2.3 may not strictly require the polynomial rate (only that the resolvent is bounded on the contour $partial B(lambda_0, gamma h^sigma)$, which (
  2. Corollary 1.16 provides), the authors should clarify which properties of $mathcal{P}$ are actually needed for the main theorem and whether Proposition 1.15(ii) is load-bearing. If it is not needed for Theorem 1.5, this should be stated; if it is, the proof gap should be addressed.
minor comments (7)
  1. Abstract: 'homogeneous polynomial' should be 'homogeneous function' or 'homogeneous polynomial (when $V in C^infty$)', since Assumption 1 allows $alpha_c notin mathbb{N}$ and $V in C^1$ (see the discussion following (1.6)).
  2. Example 1.3: The expression $V(x) = sin(4t) r^2 = 4x_1^3 x_2 - x_1 x_2^3$ appears to have a typo — the polynomial $4x_1^3 x_2 - x_1 x_2^3$ does not equal $r^2 sin(4t)$; the correct identity is $r^2 sin(4t) = 4x_1 x_2(x_1^2 - x_2^2)$, i.e., $4x_1^3 x_2 - 4x_1 x_2^3$. Please verify the coefficient 4 on the second term.
  3. Theorem 1.14 cites [8] as 'in preparation.' If this reference is not yet publicly available, the key asymptotic formula (1.23) and the claim that eigenvalues are simple and real should be briefly justified or the status of [8] clarified, since these properties are used in Example 1.9 and Example 1.10.
  4. Section 2.4.1, Case 2.2: The construction of the global diffeomorphism $psi$ on $S^{d-1}$ is technically involved. A brief remark summarizing why the non-degeneracy condition (1.5)/(1.7) is essential for this construction (ensuring $d_{theta_0} v_c neq 0$ when $v_c(theta_0) = 0$) would aid readability.
  5. Equation (1.31): The notation $f_{E,lambda,mu,s}$ has the indices in a different order than the surrounding text (which uses $f_{E,mu}$). Consistency should be fixed.
  6. The paper would benefit from a brief remark on whether full asymptotic expansions in powers of $h$ are expected when $V in C^infty$, beyond the $o(h^sigma)$ remainder (this is mentioned in passing after Remark 1.8 but could be stated more definitively).
  7. In the proof of Theorem 1.17, the application of the Gearhart–Prüss inequality via [25, Theorem 1.4] is somewhat terse; a one-sentence explanation of how the resolvent bound (1.37) translates into the semigroup decay would help readers, since this is a key output of the paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found

full rationale

The derivation chain of Theorem 1.5 is genuinely constructive and not circular. The model operator spectra Λ_c are computed independently: for signed potentials via complex dilations (Proposition 1.13, Section 3.2, using standard Aguilar–Combes theory from [1]), and for 1D monomials with n odd via Theorem 1.14 citing [8] (self-citation, but for a specific illustrative sub-case, not load-bearing for the general theorem). The parametrix construction (Proposition 2.2) connects the original operator P to the model operators P_c through scaling and cut-off functions, with error terms shown to vanish by independent scaling arguments (the identity 2−2β−σ=0 follows from the definition σ=2α/(α+2), β=2/(α+2), not from fitting). The rank argument in Section 2.3 proves Rank Π = Σ Rank Π_c by exhibiting explicit test functions and a contradiction argument, not by definition. The resolvent estimate (1.15) is derived from the parametrix, not assumed. The auxiliary operator Q = P + G² (Proposition 2.1) is analyzed via Poincaré-type inequalities and level-set geometry (Lemma 2.3), which are independent of the spectral conclusion. The scaling exponent σ = 2α/(α+2) arises from dimensional analysis of the homogeneous potential, not from matching the output. The paper is transparent about gaps (Proposition 1.4 for general non-signed V_c in d≥2), but these are honest incompleteness, not circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No invented entities. The model operators P_c are standard constructions in semiclassical analysis.

assumptions (4)
  • domain assumption Assumption 1 (Degenerate Morse): V ∈ C¹(X) real-valued; near each critical point c, ∇(V−V_c) = o(|x−c|^{α_c−1}) with V_c homogeneous of degree α_c > 1 satisfying the non-degeneracy condition ∇_x V_c(x) ≠ 0 for x ≠ c.
    This is the central structural assumption of the paper, invoked throughout Section 1.2 and used in every subsequent result. It is not a standard mathematical axiom but a domain-specific hypothesis on the potential V.
  • standard math The operator P = -h²Δ + iV with domain H²(X) and boundary conditions is maximal accretive with compact resolvent.
    Stated in Section 1.2; follows from standard accretive operator theory for C¹ real-valued V on bounded domains.
  • domain assumption Theorem 1.14: spectral asymptotics for -∂²_x + ivx^n with n odd, cited from [8] (in preparation).
    Used in Examples 1.9 and 1.10 to give explicit eigenvalue formulas for odd monomial potentials. Not load-bearing for Theorem 1.5, which only needs Proposition 1.4 (proved in the paper).
  • standard math Gearhart–Prüss type inequality with explicit bound (Theorem 1.4 of [25]).
    Used in the proof of Theorem 1.17 to convert resolvent estimates to semigroup decay estimates. Standard result in semigroup theory.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Semiclassical Schr\"odinger operators with purely imaginary potential." pith.science (2026). https://pith.science/paper/NFPMONPB

@misc{pith2026260707301,
  author       = {Pith},
  title        = {Pith review of: Semiclassical Schr\"odinger operators with purely imaginary potential},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NFPMONPB}},
  note         = {Machine review of arXiv:2607.07301}
}
abstract

We consider Schr\"odinger operators with purely imaginary potential $P = - h^{2} \Delta + i V ( x )$ on a bounded domain. Assuming that near its critical points the potential $V$ can be approximated by an homogeneous polynomial, we show that in the limit $h \to 0$ the leftmost eigenvalues of $P$ are asymptotically given by the local model associated to the most degenerated critical points of $V$. We give applications of this result to the associated evolution problem including shear flows in fluid mechanics.

Figures

Figures reproduced from arXiv: 2607.07301 by the authors.

Figure 1
Figure 1. The potential V and the eigenvalues of P in Example 1.6. Example 1.6. In X = B(0, 1) ⊂ R d , we consider the potential V (x) = |x| α , for some α > 1. In the energy surface E = 0, it has a single critical point c = 0 and V satisfies Assumption 1. Then, Theorem 1.5 and Proposition 1.13 show that, modulo o(h 2α α+2 ) terms, the eigenvalues of P near 0 are given by e i π α+2 σ(−∆ + |x| α )h 2α α+2 . This setting is ill… view at source ↗
Figure 2
Figure 2. The eigenvalues of P in Example 1.9 with d = 2, n = 4, v1 = −v2 on the left and n = 3 on the right (in filled circles). ν > 0 instead of (1.3), it should be possible to replace the o(h σ ) by a O(h σ+νe ) for some ν > e 0. For that, one should replace the big constants R, S, s−1 by some negative powers of h in the proof of Theorem 1.5. One may also ask whether the eigenvalues admit a full asymptotic expansion in fra… view at source ↗
Figure 3
Figure 3. The potential V and the eigenvalues of P in Example 1.10. part of equation (1.17), one recovers the result of Theorem 4.1.3 in [26]. It is also possible to generalize (1.16) in the spirit of (1.8) considering Vc(x) = X d j=1 [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The potential V and the eigenvalues of P in Example 1.11 with V (3)(ck) independent of k. Using hypocoercivity techniques (more precisely, considering Re⟨u, e∓iε(P − z)u⟩ with ε > 0 small enough), it has been proved in [10, Proposition 3.5] that the spectrum of P is th…
Figure 5
Figure 5. Figure 5: The energy surface V −1 (λ) near a critical point c. 2.4.1. Description of the levels sets. The first step is to understand the structure and the size of the levels set of V near E. For λ ∈ R, we recall that V −1 (λ) = {x ∈ Y ; V (x) = λ}. Following [16], we define for…
Figure 6
Figure 6. Figure 6: Some elements of the covering in Lemma 2.3 with U1, U2, U3 sat￾isfying Case 1, U4, U5 satisfying Case 2.1 and U6, U7 satisfying Case 2.2. Summing up, (2.55) for ye, (2.58), (2.60) and |xe2 − ye2| ≤ Cδ (see (2.57)) give |xe2 − λxe 1−αc 1 | ≤ |xe2 − ye2| + |ye2 − λye 1−α…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

  1. [16]

    Coti Zelati and T

    M. Coti Zelati and T. Gallay,Enhanced dissipation and Taylor dispersion in higher-dimensional parallel shear flows, J. Lond. Math. Soc. (2)108(2023), no. 4, 1358–1392

  2. [26]

    Henry,Spectre et pseudospectre d’op´ erateurs non-autoadjoints, Th` ese Universit´ e Paris Sud (2013)

    R. Henry,Spectre et pseudospectre d’op´ erateurs non-autoadjoints, Th` ese Universit´ e Paris Sud (2013). SEMICLASSICAL SCHR ¨ODINGER OPERATORS WITH PURELY IMAGINARY POTENTIAL 37

  3. [8]

    Arnaiz, J.-F

    V. Arnaiz, J.-F. Bony, and L. Michel,Asymptotic distribution of eigenvalues for a class ofPT-symmetric Schr¨ odinger operators, in preparation

  4. [1]

    Aguilar and J.-M

    J. Aguilar and J.-M. Combes,A class of analytic perturbations for one-body Schr¨ odinger Hamiltonians, Comm. Math. Phys.22(1971), 269–279

  5. [2]

    Albritton, R Beekie, and M

    D. Albritton, R Beekie, and M. Novack,Enhanced dissipation and H¨ ormander’s hypoellipticity, J. Funct. Anal.283(2022), no. 3

  6. [3]

    Almog,The stability of the normal state of superconductors in the presence of electric currents, SIAM J

    Y. Almog,The stability of the normal state of superconductors in the presence of electric currents, SIAM J. Math. Anal.40(2008), no. 2, 824–850

  7. [4]

    Almog, D

    Y. Almog, D. S. Grebenkov, and B. Helffer,Spectral semi-classical analysis of a complex Schr¨ odinger operator in exterior domains, J. Math. Phys.59(2018), no. 4

  8. [5]

    Almog, D

    Y. Almog, D. S. Grebenkov, and B. Helffer,On a Schr¨ odinger operator with a purely imaginary potential in the semiclassical limit, Comm. Partial Differential Equations44(2019), no. 12, 1542–1604

Show all 38 references
  1. [6]

    Almog and B

    Y. Almog and B. Helffer,The spectrum of a Schr¨ odinger operator in a wire-like domain with a purely imaginary degenerate potential in the semiclassical limit, M´ em. Soc. Math. Fr. (N.S.) (2020), no. 166, vi+94

  2. [7]

    Almog and R

    Y. Almog and R. Henry,Spectral analysis of a complex Schr¨ odinger operator in the semiclassical limit, SIAM J. Math. Anal.48(2016), no. 4, 2962–2993

  3. [9]

    V. I. Arnold,Local normal forms of functions, Invent. Math.35(1976), 87–109

  4. [10]

    Averseng, N

    M. Averseng, N. Frantz, F. H´ erau, and N. Raymond,Semiclassical tunneling for some 1D Schr¨ odinger operators with complex-valued potentials, arXiv:2510.04296 (2025)

  5. [11]

    Beauchard, B

    K. Beauchard, B. Helffer, R. Henry, and L. Robbiano,Degenerate parabolic operators of Kolmogorov type with a geometric control condition, ESAIM Control Optim. Calc. Var.21(2015), no. 2, 487–512

  6. [12]

    Bedrossian and M

    J. Bedrossian and M. Coti Zelati,Enhanced dissipation, hypoellipticity, and anomalous small noise inviscid limits in shear flows, Arch. Ration. Mech. Anal.224(2017), no. 3, 1161–1204

  7. [13]

    Brezis,Functional analysis, Sobolev spaces and partial differential equations, Universitext, Springer, New York, 2011

    H. Brezis,Functional analysis, Sobolev spaces and partial differential equations, Universitext, Springer, New York, 2011

  8. [14]

    Combes, P

    J.-M. Combes, P. Duclos, and R. Seiler,Kre˘ ın’s formula and one-dimensional multiple-well, J. Funct. Anal. 52(1983), no. 2, 257–301

  9. [15]

    Coti Zelati and T

    M. Coti Zelati and T. D. Drivas,A stochastic approach to enhanced diffusion, Ann. Sc. Norm. Super. Pisa Cl. Sci. (5)22(2021), no. 2, 811–834

  10. [17]

    E. B. Davies,Pseudo-spectra, the harmonic oscillator and complex resonances, R. Soc. Lond. Proc. Ser. A Math. Phys. Eng. Sci.455(1999), no. 1982, 585–599

  11. [18]

    Dencker, J

    N. Dencker, J. Sj¨ ostrand, and M. Zworski,Pseudospectra of semiclassical (pseudo-) differential operators, Comm. Pure Appl. Math.57(2004), no. 3, 384–415

  12. [19]

    Dimassi and J

    M. Dimassi and J. Sj¨ ostrand,Spectral asymptotics in the semi-classical limit, London Mathematical Society Lecture Note Series, vol. 268, Cambridge University Press, 1999

  13. [20]

    P. G. Drazin and W. H. Reid,Hydrodynamic stability, second ed., Cambridge Mathematical Library, Cambridge University Press, Cambridge, 2004, With a foreword by John Miles

  14. [21]

    Helffer,Semi-classical analysis for the Schr¨ odinger operator and applications, Lecture Notes in Mathe- matics, vol

    B. Helffer,Semi-classical analysis for the Schr¨ odinger operator and applications, Lecture Notes in Mathe- matics, vol. 1336, Springer-Verlag, 1988

  15. [22]

    Helffer,Spectral theory and its applications, Cambridge Studies in Advanced Mathematics, vol

    B. Helffer,Spectral theory and its applications, Cambridge Studies in Advanced Mathematics, vol. 139, Cambridge University Press, Cambridge, 2013

  16. [23]

    Helffer and J

    B. Helffer and J. Sj¨ ostrand,Multiple wells in the semiclassical limit. I, Comm. Partial Differential Equations 9(1984), no. 4, 337–408

  17. [24]

    Helffer and J

    B. Helffer and J. Sj¨ ostrand,From resolvent bounds to semigroup bounds, arXiv:1001.4171 (2010)

  18. [25]

    Helffer and J

    B. Helffer and J. Sj¨ ostrand,Improving semigroup bounds with resolvent estimates, Integral Equations Operator Theory93(2021), no. 3, Paper No. 36, 41

  19. [27]

    H´ erau, M

    F. H´ erau, M. Hitrik, and J. Sj¨ ostrand,Tunnel effect for Kramers-Fokker-Planck type operators, Ann. Henri Poincar´ e9(2008), no. 2, 209–274

  20. [28]

    Kato,Perturbation theory for linear operators, second ed., Grundlehren der Mathematischen Wis- senschaften, vol

    T. Kato,Perturbation theory for linear operators, second ed., Grundlehren der Mathematischen Wis- senschaften, vol. 132, Springer, 1976

  21. [29]

    Malkov,Harmonic approximation and resolvent estimates for non-self-adjoint operators, arXiv:2601.17643 (2026)

    S. Malkov,Harmonic approximation and resolvent estimates for non-self-adjoint operators, arXiv:2601.17643 (2026)

  22. [30]

    Martinez and M

    A. Martinez and M. Rouleux,Effet tunnel entre puits d´ eg´ en´ er´ es, Comm. Partial Differential Equations13 (1988), no. 9, 1157–1187

  23. [31]

    Morin, N

    L. Morin, N. Raymond, and S. V˜ u Ngo .c,Eigenvalue asymptotics for confining magnetic Schr¨ odinger operators with complex potentials, Int. Math. Res. Not. IMRN (2023), no. 17, 14547–14593

  24. [32]

    K. C. Shin,On the reality of the eigenvalues for a class ofPT-symmetric oscillators, Comm. Math. Phys. 229(2002), no. 3, 543–564

  25. [33]

    Sibuya,Global theory of a second order linear ordinary differential equation with a polynomial coefficient, North-Holland Math

    Y. Sibuya,Global theory of a second order linear ordinary differential equation with a polynomial coefficient, North-Holland Math. Stud., vol. 18, Elsevier, Amsterdam, 1975 (English)

  26. [34]

    Simon,Semiclassical analysis of low lying eigenvalues

    B. Simon,Semiclassical analysis of low lying eigenvalues. I. Nondegenerate minima: asymptotic expansions, Ann. Inst. H. Poincar´ e Sect. A (N.S.)38(1983), no. 3, 295–308

  27. [35]

    Sj¨ ostrand,Non-self-adjoint differential operators, spectral asymptotics and random perturbations, Pseudo-Differential Operators

    J. Sj¨ ostrand,Non-self-adjoint differential operators, spectral asymptotics and random perturbations, Pseudo-Differential Operators. Theory and Applications, vol. 14, Birkh¨ auser/Springer, Cham, 2019

  28. [36]

    Trefethen,Pseudospectra of linear operators, SIAM Rev.39(1997), no

    L. Trefethen,Pseudospectra of linear operators, SIAM Rev.39(1997), no. 3, 383–406

  29. [37]

    Wei,Diffusion and mixing in fluid flow via the resolvent estimate, Sci

    D. Wei,Diffusion and mixing in fluid flow via the resolvent estimate, Sci. China Math.64(2021), no. 3, 507–518

  30. [38]

    Whitney,Functions differentiable on the boundaries of regions, Ann

    H. Whitney,Functions differentiable on the boundaries of regions, Ann. of Math.35(1934), no. 3, 482–485. Universit´e de Bordeaux, Institut Math´ematique de Bordeaux Email address:victor.arnaiz@math.u-bordeaux.fr Email address:jean-francois.bony@math.u-bordeaux.fr Email address...

Pith tools

Reviewed July 9, 2026 · model on record in the stance chip above.