Universality of the transport law for supercritical quasi-periodic operators
In plain wordsThe law is not a quirk of one model; it governs a whole class of quasi-periodic systems, and for every analytic one with positive Lyapunov exponent the law for the fastest spreading follows from a single explicit hypothesis, open beyond type I.
The resultat three depths
In one sentenceThe exact law for how fast a wave spreads at its farthest in the Aubry–André model is not special to that model: for almost every phase it holds for every supercritical quasi-periodic operator whose real-analytic potential is of type I, with the least Lyapunov exponent in place of the strength of the potential, and for every real-analytic potential with positive Lyapunov exponent the paper proves the law for the fastest upper exponent conditionally on one explicit hypothesis, which is open beyond type I.
As the paper states it, in briefFor every real-analytic quasi-periodic potential that is supercritical and of type I, and almost every phase, the upper transport exponents follow the same law with replaced by the minimum Lyapunov exponent on the spectrum, for the fastest upper exponent and for every moment. Energy by energy, for packets filtered ever more narrowly near an energy , every upper moment exponent tends to , and the spectral measure has that upper local dimension at almost every . A potential obeys the law with when at frequency is supercritical and of type I. For every real-analytic potential with positive Lyapunov exponent, the law for the fastest upper exponent is proved conditionally on one explicit weak-separation hypothesis, open beyond type I.
For a scientist
One paragraph
Take with a real-analytic potential on the circle and an irrational frequency . The companion paper proved, for the almost Mathieu operator with and almost every phase, that every moment of a packet started at one site has the upper transport exponent . This paper shows that the same law holds when every energy of the spectrum is supercritical and of type I, in the sense of Avila’s global theory and of Ge and Jitomirskaya, with replaced by the minimum of the Lyapunov exponent on the spectrum: for almost every phase, the fastest upper exponent and every upper moment exponent equal , and the lower bound for holds at every phase. The law also holds energy by energy: for packets filtered ever more narrowly near an energy of the spectrum, every upper moment exponent tends to , and the spectral measure has that upper local dimension at almost every . A potential obeys the law with in place of when at frequency is supercritical and of type I. For every real-analytic potential with positive Lyapunov exponent on its spectrum, the law for follows, for almost every phase, from one explicit weak-separation hypothesis on the zeros of rational discriminants, which remains open beyond type I.
For an expert
The main theorems, as stated in the paper (Section 1)
Notation from the paper: is the operator above, with real-analytic on , irrational and ; is its spectrum, which does not depend on , and its Lyapunov exponent at energy . As in the companion paper, over the continued-fraction convergents of ; and are the tail probability beyond distance and the -th moment of the packet , averaged in time with the weight ; the upper transport exponents are and , and is the exponent of the packet started at . For , , with and .
An energy is supercritical if . Its acceleration is the right derivative at of , where is the Lyapunov exponent of the cocycle with phase complexified by ; a supercritical energy is of type I if its acceleration is one. is an explicit set of phases: those for which stays at distance at least from the turning points of the band centres of the -periodic approximants, for all large . The set of equation (6) consists of the with for all sufficiently large . is the spectral measure of , and and are the minimum and the maximum of on its support; is the upper local dimension, and and are its -essential supremum and infimum.
For a convergent , is the trace of the transfer matrix over one period of the -periodic approximant at the phase . WXS (weak separation) holds if for every and , for almost every and all large with , no energy within of the spectrum has two zeros of , real or complex and counted with multiplicity, within of each other with one of them within of .
Let be real-analytic and irrational, and assume that every energy in is supercritical and of type I. Let .
If , then for every in a set of full measure,
read as when , and for every and . When is even, contains the explicit set of (6). When the equality holds for every phase.
If , then for every phase.
The lower bound holds for every phase.
Assume the hypotheses of Theorem A with , write , and let .
For every , . If , then attains on , and for every a fixed positive fraction of the packet lies beyond distance at the times , for all large with .
Let , , and let be real, supported in , with . Then the packet started at satisfies, for every ,
for -almost every . Consequently and .
Let be real-analytic and nonconstant, let be the greatest common divisor of the frequencies with , and write . If is supercritical and of type I, then for almost every ,
read as when , and for every if . Moreover .
Let be real-analytic with positive Lyapunov exponent on its spectrum, and write as in Theorem C. If WXS holds, then for almost every ,
read as when , where is the minimum of the Lyapunov exponent on the spectrum.
By Theorems A and C, the hypothesis of Theorem D is needed only when the primitive part has acceleration at least two at some point of the spectrum; the paper does not know whether it holds there. The almost Mathieu operator with is supercritical and of type I at every spectral energy, so Theorem A gives the companion paper’s law for at every ; the companion paper proves that case independently, by a different route.
The question
The almost Mathieu operator has a single cosine for its potential, and many of its exact results lean on that special form. Is the exact transport law of the companion paper a feature of the cosine, or a law of quasi-periodic systems?
The law across a classsupercritical, type I
For every real-analytic quasi-periodic potential that is supercritical and of type I on its spectrum, in the sense of Avila’s global theory and of Ge and Jitomirskaya, the same law holds for almost every phase, with the logarithm of the coupling replaced by the minimum of the Lyapunov exponent on the spectrum. It holds for the fastest upper exponent and for the upper exponent of every moment, and the lower bound for the fastest upper exponent holds at every phase.
It also holds energy by energy. For packets filtered ever more narrowly near an energy of the spectrum, every upper moment exponent tends to , and the spectral measure has that upper local dimension at almost every ; for a filter of fixed width the paper gives two-sided bounds, set by the range of the Lyapunov exponent on the filter. The fastest spreading comes from the energies with the smallest Lyapunov exponent.
For the almost Mathieu operator this is the companion paper’s law. Beyond it, the law says the same thing about a whole class of systems: in the supercritical regime, whether a wave spreads, and how fast at its farthest, is set by how well the frequency is approximated by fractions, weighed against the Lyapunov exponent.
The frequency that counts
A potential of the form repeats times around the circle, and sampling it at step is sampling at step . The law sees this: when at frequency is supercritical and of type I, the fastest upper exponent obeys the law for almost every phase with in place of , so the arithmetic that decides transport is that of the frequency the potential actually feels. Since , the two can differ. Such a is never itself of type I.
The open frontierweak separation
For every real-analytic potential with positive Lyapunov exponent on its spectrum, type I or not, the paper shows that the law for the fastest upper exponent follows, for almost every phase, from one explicit hypothesis, weak separation: at a fixed energy, the zeros of the rational discriminants of the potential’s primitive part, as functions of the phase, do not cluster at an exponential rate near the drifting phase. The hypothesis is needed only beyond type I, for primitive potentials with acceleration at least two, and there it is open. A proof of it would give the law for the fastest upper exponent for every real-analytic quasi-periodic operator with positive Lyapunov exponent on its spectrum. Without any hypothesis, a third paper, Worst-case transport and light cones for quasi-periodic operators, proves an upper law at every phase for every such potential, in which the worst phase divides the localization strength by at most twice the acceleration.
For the two-harmonic family the paper reduces the hypothesis further, to a large-deviation estimate for one explicit polynomial, which it also leaves open. The law could also hold where weak separation fails; the paper’s methods do not decide this. The question is listed as question 50.
Questionson the index
- Partly answered Does the dimension bound of Jitomirskaya, Liu and Tcheremchantsev extend, with replaced by the minimal Lyapunov exponent, to the other cases where a sharp arithmetic transition is known? posed by others
- Gap partly closed Is this transport law universal?
- Open Does the weak-separation hypothesis hold beyond type I? left open
How this was checked
- Blind checks
- Each step independently blind-checked
- Blind audits
- A blind audit of the whole paper with no critical findings, and a blind fix-check.
- Lean
- None.
- Peer review
- Not yet.
Each check, what it covers, and what it leaves to the written proof.
| Check | Covers | Does not cover |
|---|---|---|
| Blind step checks | Each step of the argument checked independently by a reader that had not seen it. | Errors every reader shares. A check is evidence, not a guarantee. |
| Paper audit | A blind audit of the whole paper with no critical findings, and a blind fix-check. | Errors every reader shares. |
| Lean | Nothing is formalized. | |
| Peer review | Not yet. | A preprint on Zenodo. |
No computation is used in any proof. Every audit here is internal. How we check results.
Label checked .
Novelty and credit
- Novelty
- Extends named prior work: Avila’s global theory, Ge and Jitomirskaya’s type I class and the companion paper’s law for the almost Mathieu operator, credited; for general supercritical analytic potentials it credits the case to Jitomirskaya and Mavi.
The classification of energies as subcritical, critical and supercritical, and the acceleration, come from Avila’s global theory of one-frequency Schrödinger operators; the type I class, and sharp arithmetic transitions for it, are Ge and Jitomirskaya’s. For the almost Mathieu operator, the arithmetic spectral transition is Avila, You and Zhou’s; the case of the transport law is Damanik and Tcheremchantsev’s and the case Jitomirskaya and Zhang’s; the exact value in between is determined in the companion paper. For general supercritical analytic potentials, the endpoint was observed by Jitomirskaya and Mavi.
Near the transition, the paper credits lower bounds of the form to Jitomirskaya and Zhang and to Haeming, who also proved lower bounds for every moment, at every phase and along sequences of times; upper bounds on fractal dimensions of spectral measures to Jitomirskaya, Liu and Tcheremchantsev, which Levi extended, with the same factor two, to potentials that are monotone on the period; exponential dynamical localization in phase expectation for small to Jitomirskaya and Krüger; and refined growth rates for weakly Liouville frequencies to Bradshaw, de Jong, Liu, Wang, Wang and Yang.
The structural inputs are Avila’s global theory, Ge and Jitomirskaya’s type I class, the finite harmonic structure of rational discriminants of Jitomirskaya and Marx and of Shamis, and, for the spectral consequences, the inequality of Barbaroux, Germinet and Tcheremchantsev between transport and generalized dimensions. The proofs also use the complex-energy method of Damanik and Tcheremchantsev and the periodic magic formula of Damanik, Killip and Simon.
Versionsone, each with its own DOI
| Version | Date | DOI | Notes |
|---|---|---|---|
| 1.0.0 | 10.5281/zenodo.23830333 | PDF hosted here. |
Paper, code and links
papers/transport-universality-v1.0.0.pdf, version 1.0.0, 1,001,404 bytes, 55 pages.SHA-256 ae5e41bae01c014ba5b3b46a1bc2969fd0a7e68b9f7ff63e9d7a2d1939c34569
Of
paper/main.pdf, 1,001,404 bytes, in the Zenodo archive; identical in the repository at that commit. The same file is hosted here:papers/transport-universality-v1.0.0.pdf.- DOI
10.5281/zenodo.23830333; all versions 10.5281/zenodo.23830332
- Commit
5a47b0698e79155cac6c64b79bc4ffe6bf23a1b4
The commit archived by Zenodo (the repository, tag v1.0.0).
- Verify
The paper builds with pdflatex and BibTeX from the released source in
paper/of the repository; no computation is used in any proof.- Lean, in detail
None.
- Licence
Paper, LaTeX source and documentation CC BY 4.0.
Page checked against the paper, the release repository and the Zenodo record on .
Corrections
None so far.
All corrections, and what counts as one.
Cite
@misc{douglas2026universality,
author = {Douglas, Seth and Mghirbi, Nidhal},
title = {Universality of the transport law for supercritical quasi-periodic operators},
year = {2026},
version = {1.0.0},
doi = {10.5281/zenodo.23830333},
url = {https://doi.org/10.5281/zenodo.23830333},
note = {Preprint}
}