REVIEW 3 major objections 5 minor 25 references
On harmonic quasiregular mappings in Bergman spaces
T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read This paper proves that a harmonic $K$-quasiregular mapping's real and imaginary parts always lie in the same harmonic Bergman space $a^p$, for every $p>0$.
desk verdict Theorem 1 is likely right and genuinely extends the Bergman conjugate theorem to harmonic quasiregular maps for all p>0; the manuscript has a false lemma as stated and a thinly cited growth estimate, both repairable. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The argument is carried by three mechanisms. The first is the Fefferman–Stein subharmonic mean-value inequality and a gradient estimate, which convert local averages of $|u|^p$ into pointwise control of $|\nabla u|$ and hence of $|h'|$, where $f=h+g$ is the canonical decomposition into analytic parts. The second is quasiregularity itself, expressed as the analytic dilatation bound $|\omega|\le k<1$, equivalently $|g'|\le k|h'|$; this lets the estimates for $h'$ dominate the whole mapping. The third is a Möbius-invariant measure $d\tau(z)=(1-|z|^2)^{-2}dA(z)$ together with pseudo-hyperbolic disks $B_\varepsilon(a)$, which turn the local estimate into the global bound $\int_{\mathbb{D}}|h'|^p(1-|a|^2)^p\,dA(a)\le C_{p,K}\int_{\mathbb{D}}|u|^p\,dA$, and Lemma 2 reduces $\int_{\mathbb{D}}|f|^p\,dA$ to that same expression. For the univalent results, the order $\alpha=\sup_{f\in S_H}|h''(0)/2|$ and the coefficient-growth estimates (10)–(11) combine with a Hardy–Stein identity to control the integral means of $h'$.
What would settle it
Check the lemma exactly as stated with the conformal (hence $K$-quasiregular) map $f(z)=M+z$, so $h(z)=M+z$ and $g=0$: the left side $\int_{\mathbb{D}}|M+z|^p\,dA$ grows like $|M|^p$ while the right side is bounded independently of $M$ because $h'=1$, so the lemma cannot hold without the normalization $h(0)=0$ that the theorem's proof quietly uses. To test Theorem 1 itself, one would need a harmonic $K$-quasiregular $f=u+iv$ with $u\in a^p$ and $v\notin a^p$ for some $0<p\le1$; the paper offers none, and finding one would refute the theorem.
Extended reading notes
Core claim
Quasiregularity is the geometric condition that restores a Riesz-type symmetry in Bergman spaces: for a harmonic $K$-quasiregular $f=u+iv$ with $u\in a^p$, the imaginary part $v$ is automatically in $a^p$, and the norms are comparable, $\|v\|_p \le C_{p,K}\|u\|_p$, for every $0<p<\infty$. The proof treats $p>1$ by the known Hardy-space Riesz theorem for quasiregular mappings and $0<p\le1$ by a new argument using local estimates and a Möbius-invariant measure. In the second part, every univalent harmonic mapping $f$ in the normalized class $S_H$ belongs to $a^p$ for $p<1/\alpha$, where $\alpha$ is the order of the class, and $f_z, f_{\bar z}\in A^q$ and $f_\theta, r f_r\in a^q$ for $q<1/(\alpha+1)$; the same conclusions hold for $K$-quasiconformal maps with $\alpha$ replaced by $\alpha_K$.
Load-bearing premise
The proof of Theorem 1 depends on a quoted Bergman inequality for analytic $h$ that is false unless $h(0)=0$; the proof secretly has this normalization because it sets $f(0)=0$, but the lemma itself does not state it, and the inequality as written is refuted by $h(z)=M+z$ for large $M$, while Theorems 2 and 3 additionally rely on a growth estimate for $|h'|$ cited without proof.
Editorial extensions
If this is right
- If $u$ is integrable to power $p$ on the disk for a harmonic quasiregular map, so is $v$, with a norm constant depending only on $p$ and the quasiregularity constant $K$.
- For $0<p\le1$, the integral means satisfy $M_p(r,f)\le C_{p,K}(1-r)^{-1/p}\|u\|_p$, giving the same growth rate as for a single Bergman function.
- The normalized univalent harmonic class $S_H$ is contained in $a^p$ for $p<1/\alpha$, going beyond the range $p<1/\alpha^2$ that follows from the known Hardy-space result.
- The partial derivatives $f_z$, $f_{\bar z}$, $f_\theta$, and $r f_r$ of univalent harmonic maps lie in Bergman spaces for a positive range of exponents, a conclusion that typically fails in Hardy spaces.
- For $K$-quasiconformal maps in $S_H(K)$, the same membership holds with $\alpha_K$ in place of $\alpha$, and for large $K$ this beats the earlier range $p<1/(2K)$.
Reading between the lines
- The same local-estimate mechanism should extend to quasiregular mappings of the unit ball in higher dimensions, since the Fefferman–Stein inequality and pseudo-hyperbolic geometry have higher-dimensional analogues; the paper does not assert this.
- Because the quoted analytic inequality fails without the $h(0)=0$ normalization, a fully general version of Lemma 2 would need an added $|h(0)|^p$ term on the right, which would make the proof independent of the normalization $f(0)=0$.
- If the sharp Hardy-space range $p<1/\alpha$ for univalent harmonic maps is eventually established in full, the Bergman inclusion in Theorem 2 would match it, suggesting that $\alpha$ captures the real geometric obstruction; the paper proves inclusion only, not sharpness.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies analogues of the Hardy-Littlewood theorem for harmonic Bergman spaces in the class of harmonic K-quasiregular mappings of the unit disk. Theorem 1 asserts that if f=u+iv is harmonic K-quasiregular and u belongs to the harmonic Bergman space a^p, then v belongs to a^p with a norm bound depending only on p and K; the case 0<p≤1 is the main novelty. Corollary 1 gives an integral-mean growth estimate under the same hypothesis. Theorems 2 and 3 turn to univalent harmonic mappings in the classes S_H and S_H(K): they claim f∈a^p for p<1/α (resp. p<1/α_K), and that f_z, f_{\bar z}∈A^q with f_θ, r f_r∈a^q for q<1/(α+1) (resp. 1/(α_K+1)). The proofs use the Fefferman-Stein local L^p estimate, a gradient bound, Möbius invariance of the measure dτ(z)=(1-|z|^2)^{-2}dA(z), a derivative-Bergman inequality, and coefficient/growth estimates for univalent harmonic mappings.
Significance. Assuming the repaired statements, the results are significant and timely. The normalized version of Theorem 1 would close the range 0<p≤1 that is inaccessible to the Hardy-space Riesz theorem for quasiregular mappings, matching the classical Hardy-Littlewood Bergman-theorem phenomenon. The Bergman-space membership of S_H for p<1/α and the derivative conclusions for f_θ and r f_r are new and cannot be obtained from the known Hardy-space inclusions; the explicit range q<1/(α+1) for the derivatives is a concrete quantitative contribution. The proofs are based on standard tools and are mostly transparent, although the key growth estimate for h′ is quoted rather than proved.
major comments (3)
- [Theorem 1 and Corollary 1] Theorem 1 as stated is false because no normalization on f(0) (or, equivalently, on v(0)) is imposed. For any M>0 and ε>0, the function f(z)=iM+ε z is a harmonic 1-quasiregular mapping with u(z)=ε Re z and v(z)=M+ε Im z. Then u∈a^p with ∥u∥_p = ε∥Re z∥_p, while ∥v∥_p is bounded below by a positive constant times M for M large relative to ε. Hence ∥v∥_p/∥u∥_p is unbounded as M/ε→∞, contradicting the claimed inequality ∥v∥_p≤C_{p,K}∥u∥_p. The proof's first line 'Without loss of generality, we assume that f(0)=0' is not valid for the norm inequality, since subtracting f(0) changes v by the constant v(0), and v(0) is not controlled by ∥u∥_p. The theorem must be restated with v(0)=0 (or with the conclusion ∥v-v(0)∥_p≤C_{p,K}∥u∥_p). Corollary 1 inherits the same defect.
- [Section 3.1, Lemma 2] Lemma 2 is false as stated. For f≡1, one has h≡1 and g≡0, so the left side ∫_D |f|^p dA equals 1 while the right side is 0. The inequality for analytic h quoted from [25, p. 85] must either include a term |h(0)|^p on the left or assume h(0)=0. In the intended applications in Theorem 1 and Theorem 3 one has f(0)=0, hence h(0)=0, so the argument is repairable by adding the hypothesis f(0)=0 to Lemma 2. As written, however, the proof of Theorem 1 uses a false lemma at the step from (8) to (9), and the same lemma is invoked in the proof of Theorem 3.
- [Section 3.3, estimate (11)] The growth estimate |h′(re^{iθ})|≤(1+r)^{α_K-1}/(1-r)^{α_K+1} in (11) is load-bearing for Theorems 2 and 3, since it determines the admissible range p<1/α_K. The manuscript only says that 'an argument similar to that in [8, p. 98]' leads to the estimate, giving neither a proof nor the exact theorem in Duren's book. A similar estimate is evidently needed for the S_H case in Theorem 2. Because the exponent α_K+1 enters directly into the final range of p, the authors should state (11) as a lemma with a complete proof, or provide a precise citation, before the paper can be accepted.
minor comments (5)
- [Section 1.1 and Lemma 1] The representation of a complex-valued harmonic f is consistently written as f=h+g, but it should be f=h+\bar g with g(0)=0; the authors themselves use f=h+\bar g in Theorem 3. The current notation makes the definition of F=h+g in Lemma 1 confusing.
- [Section 3.3] In the proof of Theorem 3, the displayed identity f = 1/2(H+iG)+1/2(H-iG) is missing a conjugation: the correct formula is f = 1/2(H+iG)+1/2\overline{(H-iG)}. The subsequent identification of g with 1/2(H-iG) depends on this.
- [Section 3.3] The identities for the angular and radial derivatives should be -i f_θ = z h' - \overline{z g'} and r f_r = z h' + \overline{z g'}; the overline on z g' is missing in the text.
- [Section 3.1] The claim that the case p>1 of Theorem 1 follows immediately from Theorem C needs a short dilation argument: for u_r(z)=u(rz), Theorem C gives M_p(s,v_r)≤C M_p(s,u_r), and one integrates over s and lets r→1. A sentence explaining this would help the reader.
- [Section 3.1] The change of order of integration in (7) is justified by Tonelli's theorem since all integrands are nonnegative; a brief remark to this effect would improve readability.
Circularity Check
No circularity detected: Theorem 1 rests on external Bergman and derivative inequalities, and Theorems 2–3 use only the defining order constants α and α_K.
full rationale
The derivation chain of Theorem 1 is self-contained with respect to external theorems. Lemma 1 combines the Fefferman–Stein mean inequality (Theorem D) with a standard gradient estimate (Theorem E) to control |h'(0)|; Möbius invariance of dτ then yields estimate (8). Lemma 2 supplies the analytic Bergman derivative inequality quoted from Zhu [25, p. 85], and Theorem 1 follows by combining (8) and (9). None of these steps re-uses the conclusion of Theorem 1, and the constant C_{p,K} depends only on p and K, not on any fitted data. Theorems 2 and 3 use the defining quantities α = sup |h''(0)/2| and α_K over the respective classes; these are structural constants, not fitted parameters, and the supporting estimates (10)–(11) are cited to Duren [8] and obtained by standard Möbius-invariance and Hardy–Stein arguments. The paper's self-citations ([5], [6], [7]) appear only as background context and are not load-bearing in the proofs. One non-circular correctness issue should be flagged: Lemma 2 states the analytic inequality ∫_D |h|^p dA ≤ C_p ∫_D |h'(a)|^p (1−|a|^2)^p dA(a) without the h(0)=0 normalization or the |h(0)|^p term; as stated it fails for constant h. In the applications f(0)=0 (and h(0)=g(0)=0 in S_H), so h(0)=0 and the intended inequality applies, making the gap repairable. This is a mathematical misstatement, not a circularity.
Assumptions & free parameters
assumptions (7)
- standard math Fefferman-Stein local mean inequality (Theorem D): |u(a)|^p ≤ C_p / |D_r(a)| ∫_{D_r(a)} |u|^p dA for harmonic u.
- standard math Gradient estimate for harmonic functions (Theorem E): |∇u(a)| ≤ K r^{-1} sup_{D_r(a)} |u|.
- standard math Bergman derivative characterization: ∫_D |h|^p dA ≤ C_p (|h(0)|^p + ∫_D |h'|^p (1-|z|^2)^p dA) for analytic h.
- standard math Möbius invariance of the measure dτ(z) = (1-|z|^2)^{-2} dA(z).
- standard math Hardy-Stein identity for analytic functions.
- domain assumption Growth estimate (11): |h'(re^{iθ})| ≤ (1+r)^{α_K-1}/(1-r)^{α_K+1} for f ∈ S_H(K).
- domain assumption The automorphism construction F(z) = [f(φ_ζ(z)) - f(ζ)] / [(1-|ζ|^2) h'(ζ)] lies in S_H(K).
Cite this review
Pith. "Pith review of On harmonic quasiregular mappings in Bergman spaces." pith.science (2026). https://pith.science/paper/VSA73AHG
@misc{pith2026250719158,
author = {Pith},
title = {Pith review of: On harmonic quasiregular mappings in Bergman spaces},
year = {2026},
howpublished = {\url{https://pith.science/paper/VSA73AHG}},
note = {Machine review of arXiv:2507.19158}
}
abstract
A classical result of Hardy and Littlewood says that if $f=u+iv$ is analytic in the unit disk $\mathbb{D}$ and $u$ is in the harmonic Bergman space $a^p$ ($0<p<\infty$), then $v$ is also in $a^p$. This complements a celebrated result of M. Riesz on Hardy spaces, which only holds for $1<p<\infty$. These results do not extend directly to complex-valued harmonic functions. We prove that the Hardy-Littlewood theorem holds for a harmonic function $f=u+iv$ if we place the assumption that $f$ is quasiregular in $\mathbb{D}$. This makes further progress on the recent Riesz type theorems for harmonic quasiregular mappings by several authors. Then we consider univalent harmonic mappings in $\mathbb{D}$ and study their membership in Bergman spaces. In particular, we produce a non-trivial range of $p>0$ such that every univalent harmonic function $f$ (and the partial derivatives $f_\theta,\, rf_r$) is of class $a^p$. This result extends nicely to harmonic quasiconformal mappings in $\mathbb{D}$.
Reference graph
Works this paper leans on
-
[1]
Y. Abu-Muhanna and A. Lyzzaik. The boundary behaviour of harmonic univalent maps. Pacific J. Math. , 141(1):1–20, 1990
work page 1990
-
[2]
K. Astala and P. Koskela. Hp-theory for quasiconformal mappings. Pure Appl. Math. Q. , 7(1):19–50, 2011
work page 2011
-
[3]
Axler, P
S. Axler, P. Bourdon, and W. Ramey. Harmonic function theory , volume 137 of Graduate Texts in Mathematics. Springer-Verlag, New York, second edition, 2001
2001
-
[4]
S. Chen and M. Huang. Riesz type theorems for κ-pluriharmonic mappings, invariant har- monic quasiregular mappings and harmonic quasiregular mappings.arXiv:2310.15452, 2023
arXiv 2023
-
[5]
S. Das, J. Huang, and A. Rasila. Zygmund’s theorem for harmonic quasiregular mappings. Complex Anal. Oper. Theory , 19:Article no. 91, 13 pp., 2025
2025
-
[6]
Note on real and imaginary parts of harmonic quasiregular mappings
S. Das and A. Rasila. Note on real and imaginary parts of harmonic quasiregular mappings. arXiv:2506.04618, 2025
work page Pith review arXiv 2025
-
[7]
Das and A
S. Das and A. Sairam Kaliraj. Integral mean estimates for univalent and locally univalent harmonic mappings. Canad. Math. Bull. , 67(3):655–669, 2024
2024
-
[8]
P. Duren. Harmonic mappings in the plane, volume 156 ofCambridge Tracts in Mathematics. Cambridge University Press, Cambridge, 2004
work page 2004
Show all 25 references
-
[9]
Duren and A
P. Duren and A. Schuster. Bergman spaces, volume 100 of Mathematical Surveys and Mono- graphs. American Mathematical Society, Providence, RI, 2004
2004
-
[10]
P. L. Duren. Theory ofHp spaces. Pure and Applied Mathematics, Vol. 38. Academic Press, New York-London, 1970
1970
-
[11]
Fefferman and E
C. Fefferman and E. M. Stein. Hp spaces of several variables. Acta Math., 129(3-4):137–193, 1972
1972
-
[12]
G. H. Hardy and J. E. Littlewood. Some properties of conjugate functions. J. Reine Angew. Math., 167:405–423, 1932. On harmonic quasiregular mappings in Bergman spaces 13
1932
-
[13]
Hedenmalm, B
H. Hedenmalm, B. Korenblum, and K. Zhu. Theory of Bergman spaces , volume 199 of Graduate Texts in Mathematics . Springer-Verlag, New York, 2000
2000
-
[14]
D. Kalaj. Riesz and Kolmogorov inequality for harmonic quasiregular mappings. J. Math. Anal. Appl., 542(1):Paper No. 128767, 15 pp., 2025
2025
-
[15]
D. Kalaj. Zygmund theorem for harmonic quasiregular mappings. Anal. Math. Phys. , 15(2):Paper No. 41, 13 pp., 2025
2025
-
[16]
Kalaj and R
D. Kalaj and R. Meˇ strovi´ c. Isoperimetric type inequalities for harmonic functions.J. Math. Anal. Appl., 373(2):439–448, 2011
2011
-
[17]
Li and S
P. Li and S. Ponnusamy. On the coefficients estimate of K-quasiconformal harmonic map- pings. arXiv:2504.08284, 2025
2025
-
[18]
Liu and J.-F
J. Liu and J.-F. Zhu. Riesz conjugate functions theorem for harmonic quasiconformal map- pings. Adv. Math., 434:Paper No. 109321, 27, 2023
2023
-
[19]
M. Nowak. Integral means of univalent harmonic maps.Ann. Univ. Mariae Curie-Sk lodowska Sect. A, 50:155–162, 1996
1996
-
[20]
Pavlovi´ c.Function classes on the unit disc—an introduction , volume 52 of De Gruyter Studies in Mathematics
M. Pavlovi´ c.Function classes on the unit disc—an introduction , volume 52 of De Gruyter Studies in Mathematics . De Gruyter, Berlin, 2nd edition, 2019
2019
-
[21]
Pommerenke
C. Pommerenke. Univalent functions. Vandenhoeck & Ruprecht, G¨ ottingen,,, 1975. With a chapter on quadratic differentials by Gerd Jensen
1975
-
[22]
Sobolewski
P. Sobolewski. Properties of harmonic conjugates. Ann. Univ. Mariae Curie-Sk lodowska Sect. A, 62:143–147, 2008
2008
-
[23]
Wang, X.-Y
Z.-G. Wang, X.-Y. Wang, A. Rasila, and J.-L. Qiu. On a problem of Pavlovi´ c involving harmonic quasiconformal mappings. arXiv:2405.19852, 2024
2024
-
[24]
Watanabe
H. Watanabe. Some properties of functions in Bergman space Ap. Proc. Fac. Sci. Tokai Univ., 13:39–54, 1978
1978
-
[25]
K. Zhu. Operator theory in function spaces, volume 138 of Mathematical Surveys and Mono- graphs. American Mathematical Society, Providence, RI, second edition, 2007. Suman Das Department of Mathematics with Computer Science, Guangdong Technion - Israel Institute of Technology,...
2007
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.