Vectorial Kato inequality for p p -harmonic maps with optimal constant
Andreas Gastel, Katarzyna Mazowiecka, Michał Miśkiewicz · 2025
Read Vectorial Kato inequality for p p -harmonic maps with optimal constant by Andreas Gastel, Katarzyna Mazowiecka, Michał Miśkiewicz (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-05038.txt.
Introduction
Andreas Gastel, Katarzyna Mazowiecka, and Michał Miśkiewicz prove a sharp bound for bent field maps in a 2025 paper. The bound caps how sharp a map's slope can run next to its bulk pull. Its best scale for map forms parts ways with the known one-value case, they show.
That split is a shock next to the old lore of the field at hand. The bound then feeds checks of how smooth such maps must be in their zones. For ball-to-sphere maps, a wider span of the bent grade now earns smooth repute.
Catalog record
| CREATOR | Andreas Gastel, Katarzyna Mazowiecka, Michał Miśkiewicz |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-05038
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Vectorial Kato inequality for p p -harmonic maps with optimal constant Andreas Gastel, Katarzyna Mazowiecka, Michał Miśkiewicz Abstract . We derive the sharp vectorial Kato inequality for p p -harmonic mappings. Surprisingly, the optimal constant differs from the one obtained for scalar valued p p -harmonic functions by Chang, Chen, and Wei [ 2 ] . As an application we demonstrate how this inequality can be used in the study of regularity of p p -harmonic maps. Furthermore, in the case of p p -harmonic maps from B 3 B^{3} to 𝕊 3 {\mathbb{S}}^{3} , we enhance the known range of p p values for which regularity is achieved. Specifically, we establish that for p ∈ [ 2 , 2.642 ] p\in[2,2.642] , minimizing p p -harmonic maps must be regular. 2010 Mathematics Subject Classification 58E20, 35B65, 35J60 1. Introduction For differentiable vector fields, or sections of a vector bundle with metric connection \nabla , the trivial pointwise inequality |\nabla V|\geq|\nabla|V|| is known as Kato’s inequality. It has been known for a long time that the constant 1 in this inequality can be improved for solutions of (systems of) partial differential equations. This dates back at least as early as 1975 when Schoen, Simon, and Yau [ 12 ] proved |\nabla A|^{2}\geq\frac{n+2}{n}\,|\nabla|A||^{2} for the second fundamental form A of any minimal hypersurface in \mathbb{R}^{n+1} . This may not have been stated explicitly in [ 12 ] , see [ 1 , Lemme 3] for this interpretation. For harmonic maps u\colon\mathcal{M}^{n}\to\mathcal{N}^{d} , the vector field to be considered is \nabla u , giving estimates between |\nabla^{2}u| and |\nabla|\nabla u|| . The most striking application of Kato inequalities here is Okayasu’s paper [ 11 ] . He proved |\nabla|\nabla u||^{2}\leq\frac{n-1}{n}\,|\nabla^{2}u|^{2} in the sense of distributions, with optimal constant, and used this to improve the results by Schoen and Uhlenbeck [ 13 ] about partial regularity for harmonic maps to spheres, resulting in optimal regularity for maps to {\mathbb{S}}^{4} and {\mathbb{S}}^{5} . Note, however, that Schoen and Uhlenbeck already had some improved (but non-optimal) Kato-type inequality with constant \frac{2nd}{2nd+1} instead of \frac{n-1}{n} . Improved Kato type inequalities are most significant in regularity proofs where the values of constants are crucial. This is the case in all of the papers mentioned so far, and more generally in many regularity proofs that perform “dimension reduction of the singular set” by ruling out conical solutions — representing the blow-up of a solution near a singular point — by playing out a stability inequality against some Bochner–Weitzenböck formula. As this technique has widespread applications, the search for optimal constants in Kato’s inequality is relevant for solutions of many types of equations and systems. For harmonic maps to targets of any dimension, Okayasu’s constant \frac{n-1}{n} is optimal. The situation is different for p -harmonic maps, which are defined as critical points of (1.1) E_{p}(u)=\int_{\mathcal{M}}|\nabla u|^{p}\qquad\text{among $u\in W^{1,p}(\mathcal{M},\mathcal{N})$.} It is remarkable that the optimal constant for p -harmonic maps is not the same in the scalar and the vectorial case. This may be one of the reasons why Kato-type inequalities for p -harmonic maps have not been proven but recently. Let us write the inequalities in the form |\nabla^{2}u|^{2}\geq\kappa|\nabla|\nabla u||^{2} here. In 2016, Chang, Chen, and Wei [ 2 ] proved a Kato inequality for p -harmonic functions ( \mathbb{R} -valued) with the optimal constant \kappa=\min\{1+\frac{(p-1)^{2}}{n-1},2\} . In 2019, the first-named author [ 6 ] studied a micropolar elasticity model which couples p -harmonic maps with some other equation. In the course of a proof that the theory is sort of well-behaved also for exponents p slightly larger than 2 , he provided a first Kato-type inequality for targets other than \mathbb{R} , with some non-optimal \kappa>1 for suitable combinations of n and p . This prompted the second- and third-named authors [ 8 ] in 2023 to study the n=2 case more thoroughly, with the optimal result \kappa=\min\{1+(p-1)^{2},2\} , which happens to coincide with the constant for scalar-valued case mentioned above. These results, combined with the harmonic case p=2 , might lead one to guess that \kappa=\min\{1+\frac{(p-1)^{2}}{n-1},2\} could be the best constant in the general case n\geq 2 , p>1 . But [ 8 ] also gave numerical evidence that in some cases the constant should be worse than that. This was our motivation to tackle the problem of the optimal Kato constant for p -harmonic maps in general. Our result is that the constant is given by \kappa(p,n)=\left\{\begin{array}[]{lll}1+\frac{(p-1)^{2}}{n-1}&\quad\text{for }&1\leq p\leq 1+\frac{n-1}{\sqrt{2n}-1}\\ 2-\frac{(p-\sqrt{2n})^{2}}{(\sqrt{n}-\sqrt{2})^{2}}&\quad\text{for }&1+\frac{n-1}{\sqrt{2n}-1}\leq p\leq\sqrt{2n}\\ 2&\quad\text{for }&p\geq\sqrt{2n}.\end{array}\right. The constant above is sharp in a pointwise sense: it is achieved by some map that solves the p -harmonic equation at the given point. This is the main result of the present paper and it is proved in Section 2 . p \kappa(p,n) 1 1+\tfrac{n-1}{(\sqrt{2n}-1)^{2}} \sim 3/2 2 1 1+\tfrac{n-1}{\sqrt{2n}-1} \sim\sqrt{n/2} \sqrt{2n} \kappa=1+\tfrac{(p-1)^{2}}{n-1} \kappa=2-\tfrac{(p-\sqrt{2n})^{2}}{(\sqrt{2n}-1)^{2}} \kappa=2 Figure 1. A graph showing the dependence of \kappa(p,n) on p (assuming d\geq 2 ). The dashed curve illustrates the best constant \kappa(p,n) in the scalar case d=1 [ 2 ] . Note that the shape presented above corresponds to the asymptotic case n\to\infty , where the first two regimes have the same length. For small n , the middle regime can be very little (its length is \approx 0.07 for n=3 ) or even absent (for n=2 , as found in [ 8 ] ). In general, \kappa(p,n) is a piecewise quadratic function of class C^{1} . In Section 3 we show how our Kato type inequality can be used to study the regularity of p -harmonic maps. As already mentioned in this introduction, the reasoning is classical and follows by combining a stability inequality with a Bochner formula, as in [ 13 ] . For suitably chosen p , n and d , this leads to full or partial regularity of minimizing p -harmonic maps B^{n}\to{\mathbb{S}}^{d} . In particular, if n and d are chosen so that harmonic maps from {B}^{n} into {\mathbb{S}}^{d} are known to be regular (see [ 8 ] for a summary of all such cases), then we provide an interval of p\in[2,p_{0}(n,d)] for which p -harmonic maps are also regular. To give a concrete example, we study the case of p -harmonic maps B^{4}\to S^{4} in detail, obtaining the interval p\in[2,p_{0}] with p_{0}=\frac{5+\sqrt{13}}{3}\approx 2.86 . Section 4 provides further improvement in the special case of p -harmonic maps B^{3}\to{\mathbb{S}}^{3} , which was the main focus in [ 8 ] . There regularity was established for all 2\leq p\leq 2.366 and p\geq 2.962 . Here, the main innovation lies in a new choice of test functions for both the stability and Bochner inequalities: instead of |\nabla u| and |\nabla u|^{p-2} (respectively), which is the natural choice, one can take |\nabla u|^{1+\gamma} and |\nabla u|^{p-2+2\gamma} with an additional parameter \gamma . By formal calculation, one can check that choosing \gamma strictly below zero leads to a better constant. However, an optimal choice \gamma\approx-0.6 is not admissible, as the exponent p-2+2\gamma becomes negative in this case. For this reason, we perform a non-trivial regularization procedure, and finally obtain regularity of minimizing p -harmonic maps B^{3}\to{\mathbb{S}}^{3} for all 2\leq p\leq 2.642 as a result. Acknowledgments. The project is co-financed by: • the Deutsche