Well-posedness of half-harmonic map heat flows for rough initial data
Kilian Koch, Christof Melcher · 2025
Read Well-posedness of half-harmonic map heat flows for rough initial data by Kilian Koch, Christof Melcher (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-2504-06933.txt.
Introduction
Kilian Koch and Christof Melcher widen the class of starts for which a heat-like map flow is well behaved. The flow is the half-harmonic map heat flow, a nonlocal cousin of a classic shape-smoothing flow. It has lately been tied to free boundary minimal surfaces. Their 2025 paper borrows a theory built for the Navier-Stokes equations by Koch and Tataru. That theory rests on Carleson measure checks.
Within it, the pair set up a sharp, scale-matched frame for rough starts. The frame admits starts rougher than plain energy or Sobolev classes allow. For those starts they prove a solution exists, is the only one, and shifts smoothly as the start shifts. Even spread-out starts that spawn self-similar growth fit inside. The gain is a wider, safer base for flow studies.
Catalog record
| CREATOR | Kilian Koch, Christof Melcher |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-06933
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Well-posedness of half-harmonic map heat flows for rough initial data Thanks: Kilian Koch (corresponding author): Lehrstuhl für Angewandte Analysis, RWTH Aachen University, Kreuzherrenstr. 2, 52062 Aachen, Germany. Email: [email protected] . Phone: +49 241 80-94586. Thanks: Christof Melcher: Lehrstuhl für Angewandte Analysis, RWTH Aachen University, Kreuzherrenstr. 2, 52062 Aachen, Germany. Email: [email protected] . Phone: +49 241 80-94585. Kilian Koch, Christof Melcher Abstract . We adopt the Koch-Tataru theory for the Navier-Stokes equations, based on Carleson measure estimates, to develop a scaling-critical low-regularity framework for half-harmonic map heat flows. This nonlocal variant of the harmonic map heat flow has been studied recently in connection with free boundary minimal surfaces. We introduce a new class of initial data for the flow, broader than the conventional energy or Sobolev spaces considered in previous work, for which we establish existence, uniqueness, and continuous dependence. 1. Introduction Fractional harmonic maps and flows arise at the intersection of several areas of mathematics, appearing as continuum limits of certain spin systems and in connection with the Plateau problem for disc-type minimal surfaces. Fractional harmonic maps generalize classical harmonic maps by replacing the governing Dirichlet energy with a fractional version introducing non-local interactions between points at various scales, see e.g. [ 5 , 7 , 24 , 25 ] . Notably, fractional harmonic flows are governed by evolution equations that model the time evolution of maps under non-local diffusion processes. Due to their deep and unexpectedly intricate mathematical structure, fractional harmonic flows have recently attracted significant attention in the geometric analysis community, see e.g. [ 19 , 32 , 35 , 36 , 37 , 38 ] . In this work, we consider the initial value problem for half-harmonic map heat flows from euclidean spaces into spheres u:\mathbb{R}^{n}\times(0,T)\to\mathbb{S}^{m-1} , governed by the equation \partial_{t}u+(-\Delta)^{\frac{1}{2}}u\perp T_{u}\mathbb{S}^{m-1} and initial conditions a:\mathbb{R}^{n}\to\mathbb{S}^{m-1} in suitable function spaces. Here (-\Delta)^{\frac{1}{2}} denotes the usual fractional Laplacian defined, e.g., via Fourier transform. The equation corresponds to the L^{2} gradient flow of the square of the fractional Sobolev norm (1) \mathcal{E}_{\frac{1}{2}}(u)=\frac{\gamma_{n}}{4}\int_{\mathbb{R}^{n}}\int_{\mathbb{R}^{n}}\frac{|u(x)-u(y)|^{2}}{|x-y|^{n+1}}\,dxdy. where \gamma_{n}:=\Gamma((n+1)/2)/\pi^{(n+1)/2} . The abstract weak form of the flow equation reads \left\langle\partial_{t}u,\phi\right\rangle_{L^{2}}+\mathcal{E}_{\frac{1}{2}}^{\prime}(u)\langle\phi\rangle=0 for all smooth tangent fields \phi\perp u . Fundamental questions of the existence and uniqueness of half-harmonic flows from \mathbb{S}^{1} have been examined in [ 35 , 36 , 37 , 32 , 38 ] , including local well-posedness in the optimal Sobolev space H^{\frac{1}{2}} and a bubbling analysis at possible blow-up points that allows for an extension to a global weak solution in full analogy to the well-known result for the conventional harmonic map heat flow from surfaces [ 31 ] . Infinite time blow-up has indeed been established in [ 28 ] with a conjecture of non-occurance of finite time singularities for half-harmonic map heat flows as purely non-local phenomenon. Existence and partial regularity results for global weak solutions in higher dimensions were obtained in [ 26 , 13 ] . In this work, we shall examine well-posedness in more general class scale invariant spaces larger than the energy space \dot{H}^{\frac{1}{2}}(\mathbb{R};\mathbb{S}^{m-1}) , or more generally the critical Sobolev \dot{H}^{\frac{n}{2}}(\mathbb{R}^{n};\mathbb{S}^{m-1}) as examined in [ 23 , 16 ] . One motivation arises from situations which require rougher initial conditions, such as 1-homogeneous fields that involve point or jump discontinuities at the origin. These are crucial for the construction of self-similar expanders, as explored in [ 9 , 12 ] for the conventional harmonic map heat flow and the Landau-Lifshitz-Gilbert equation, respectively. Another motivation of more general character is to explore the conditions under which solutions exist or may fail to exist, as well as under what conditions solutions remain smooth or develop singularities, a well-known program in the context of the Navier-Stokes equation. A somewhat optimal framework established in [ 15 ] is to exploit the connection between heat semigroups and \mathrm{BMO} spaces through their maximal characterization in terms of Carleson measures [ 29 , 11 , 18 ] . The concept was also realized for a wide range of geometric flows [ 14 ] and other nonlinear parabolic evolution equation, see e.g. [ 2 , 12 , 27 , 21 ] , including in particular harmonic map heat flows and their coupling to fluid dynamic equations in models of liquid crystals [ 33 , 34 ] . For the harmonic map heat flow \partial_{t}u-\Delta u=|\nabla u|^{2}u admissible initial conditions a:\mathbb{R}^{n}\to\mathbb{R}^{m} are characterized via their caloric extension u:\mathbb{R}^{n+1}_{+}\to\mathbb{R}^{m} , i.e. u(t)=g_{t}\ast a where g_{t} is the heat kernel, so that, for some time horizon T>0 , local energy averages \sup_{0<r<\sqrt{T}}\sup_{x\in\mathbb{R}^{n}}\int_{0}^{r^{2}}\fint_{B_{r}(x)}|\nabla u|^{2} satisfy a certain smallness condition. This is nothing but a local Carleson condition on the energy density of u considered as a measure on \mathbb{R}^{n+1}_{+} . In the global case T=\infty it provides a homogeneous norm equivalent to the \mathrm{BMO} norm. In the half-harmonic case, we adopt a similar strategy. However, instead of the caloric extension, we consider the harmonic extension v:\mathbb{R}^{n+1}_{+}\to\mathbb{R}^{m} given by v(t)=p_{t}\ast a where p_{t} is the Poisson kernel for t>0 . Then v is the solution to the homogeneous linear problem \partial_{t}v+(-\Delta)^{\frac{1}{2}}v=0 with v(0)=a . The fractional energy density is directly derived from the expression for the fractional energy \mathcal{E}_{\frac{1}{2}}(v) in ( 1 ) (2) \left|d_{\frac{1}{2}}v\right|_{od}^{2}(x)=\frac{\gamma_{n}}{2}\int_{\mathbb{R}^{n}}\frac{|v(x+h)-v(x)|^{2}}{|h|}\frac{dh}{|h|^{n}}. The notation refers to the off-diagonal norm (see section 3.2 ) of the fractional gradient (3) d_{\frac{1}{2}}v(x,y)=\sqrt{\frac{\gamma_{n}}{2}}\frac{v(x)-v(y)}{|x-y|^{\frac{1}{2}}} introduced in [ 22 ] to capture compensated compactness and improved regularity phenomena in fractional harmonic maps. Notably, as observed and used in [ 36 ] , the half-harmonic map heat flow equation takes the explicit form (4) \partial_{t}u+(-\Delta)^{\frac{1}{2}}u=\left|d_{\frac{1}{2}}u\right|_{od}^{2}u. This motivates the following semi-norm on initial conditions a\in L^{\infty}(\mathbb{R}^{n};\mathbb{R}^{m}) (5) [a]_{\mathcal{A}_{T}}^{2}:=\sup_{0<r<T}\sup_{x\in\mathbb{R}^{n}}\int_{0}^{r}\fint_{B_{r}(x)}\left|d_{\frac{1}{2}}v\right|_{od}^{2}\quad\text{where}\quad v=p_{t}\ast a for T\in[0,\infty] to define scaling-invariant function spaces (6) \mathcal{A}=\{a\in L^{\infty}(\mathbb{R}^{n};\mathbb{R}^{m}):[a]_{\mathcal{A}_{\infty}}<\infty\} and (7) \mathcal{V}=\{a\in\mathcal{A}:\lim_{T\to 0}[a]_{\mathcal{A}_{T}}=0\}. The spaces \mathcal{A} and \mathcal{V} replace \mathrm{BMO} and \mathrm{VMO} , respectively, but strictly include the scaling-invariant Sobolev and Besov spaces, i.e. (8) L^{\infty}\cap\dot{H}^{\frac{n}{2}}(\mathbb{R}^{n};\mathbb{R}^{m})\subset L^{\infty}\cap\dot{B}^{\frac{n}{2}}_{2,\infty}(\mathbb{R}^{n};\mathbb{R}^{m})\subset\mathcal{A}. More precisely, \mathcal{A} contains the Q_{\alpha} spaces for \alpha\geq 0 as introduced in [ 6 ] . Those spaces capture fractional mean oscillation and serve as a bridge between \mathrm{BMO} and Hölder-type regularity, though they are not interpolation spaces in the classical sense. A central feature of the Q_{\alpha} framework is its characterization via Carleson measures \alpha\in(0,1) . Buiding on this characterization the well-posedness of certain fractional variants of the Navier-Stokes equation in generalized Q spaces have been investigated in [ 20 ] . Notably, the limiting case Q_{0} , captured by our results, forms a proper subspace of \mathrm{BMO} , offering a slightly more refined control of oscillation. In particular, we observe that the space \mathcal{A} permits jump discontinuities in one dimension and point defects in higher dimensions. Our key result establishes solvability of ( 4 ) for initial conditions a:\mathbb{R}^{n}\to\mathbb{S}^{m-1} in \mathcal{A} with [a]_{\mathcal{A}_{T}} sufficiently small up to time T in certain spaces X_{T} to be defined below. In particular we obtain local existence for a\in\mathcal{V} and global existence if [a]_{\mathcal{A}} is sufficiently small. Moreover, uniqueness holds in the space X_{T} , with continuous dependence on the initial data. Our approach is primarily based on real space methods, i.e, on the fractional gradient rather than the fractional Laplacian. The specific choice of the spaces \mathcal{A} and X_{T} is particularly well-suited to the structure of the nonlinearity in ( 4 ) involving