Dirac operators twisted by ramified Euclidean line bundles
Gorapada Bera, Thomas Walpuski · 2025
Read Dirac operators twisted by ramified Euclidean line bundles by Gorapada Bera, Thomas Walpuski (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-2503-01392.txt.
Introduction
Gorapada Bera and Thomas Walpuski wrote a 2025 math paper on Dirac operators twisted by special line bundles. The bundles branch along a set, a setup linked to harmonic spinors that show up in gauge theory and in the study of special shapes. The authors study which closed forms the operator can take.
They describe those closed forms through a quotient space drawn from past work on the topic. When the branch set is a smooth closed surface-like set of two fewer dimensions, they build a concrete form of that quotient. From that step, they set up a theory of smooth behavior for square integrable solutions. The paper is a piece of pure analysis with ties to geometry.
Catalog record
| CREATOR | Gorapada Bera, Thomas Walpuski |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 138 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-01392
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Dirac operators twisted by ramified Euclidean line bundles Gorapada Bera, Thomas Walpuski Abstract This article is concerned with the analysis of Dirac operators D D twisted by ramified Euclidean line bundles ( Z , 𝔩 ) (Z,{\mathfrak{l}}) —motivated by their relation with 𝐙 / 2 𝐙 \mathbf{Z}/2\mathbf{Z} harmonic spinors, which have appeared in various contexts in gauge theory and calibrated geometry. The closed extensions of D D are described in terms of the Gelfand–Robbin quotient 𝐇 ˇ \check{\mathbf{H}} . Assuming that the branching locus Z Z is a closed codimension two submanifold, a geometric realisation of 𝐇 ˇ \check{\mathbf{H}} is constructed. This, in turn, leads to an L 2 L^{2} regularity theory. \useosf Dedicated to the memory of Dietmar Salamon. 1 Introduction Let (X,g) be a closed Riemannian manifold of dimension n . Definition 1.1 (cf. [ 22 , 6 ] ) . A Dirac bundle with skew torsion on (X,g) consists of: (1) a Euclidean vector bundle S over X equipped with a skew-adjoint Clifford multiplication \gamma\mskip 0.5mu\colon\thinspace TX\to{\mathfrak{o}}(S) ; that is: \gamma(v)^{2}=-\lvert v\rvert^{2}\,\mathbf{1}_{S} for every v\in TX ; and (2) an orthogonal covariant derivative \nabla on S and a 3 –form \Tor\in\Omega^{3}(X) such that \gamma is parallel with respect to \nabla and the orthogonal affine connection \nabla^{T} on (X,g) defined by \langle\nabla_{u}^{T}v,w\rangle=\langle\nabla_{u}^{\mathrm{LC}}v,w\rangle+\tfrac{1}{2}\Tor(u,v,w). Here \nabla^{\mathrm{LC}} denotes the Levi-Civita connection of (X,g) . Definition 1.2 . A ramified Euclidean line bundle over X consists of: (1) a closed subset Z\subset X , the branching locus , and (2) a Euclidean line bundle {\mathfrak{l}} over X{\backslash}Z such that (3) if W\subset Z is closed and {\mathfrak{l}} extends over X{\backslash}W , then W=Z . This article is concerned with the analysis of the Dirac operator associated with a Dirac bundle with skew torsion (S,\gamma,\nabla,\Tor) twisted by a ramified Euclidean line bundle {\mathfrak{l}} D\mskip 0.5mu\colon\thinspace H^{1}\Gamma\lparen X{\backslash}Z,S\otimes{\mathfrak{l}}\rparen\to L^{2}\Gamma\lparen X{\backslash}Z,S\otimes{\mathfrak{l}}\rparen and its extensions. The authors’ motivation for this stems from the following. Taubes has observed that the failure of compactness for a wide variety of generalised Seiberg–Witten equations—e.g.: stable flat \mathbf{P}\mathrm{SL}_{2}({\mathbf{C}}) –connections in dimension three [ 36 ] , anti-self-dual \mathrm{SL}_{2}({\mathbf{C}}) –connections in dimension four [ 35 ] , the Seiberg–Witten equation with multiple spinors [ 37 ] , the Vafa–Witten equation [ 38 ] , and the Kapustin–Witten equation [ 39 ] — leaves behind evidence in the form of a \mathbf{Z}/2\mathbf{Z} harmonic spinor . The latter is a pair (Z,{\mathfrak{l}};\Phi) consisting of a ramified Euclidean line bundle (Z,{\mathfrak{l}}) and a harmonic spinor \Phi\in\ker D . \mathbf{Z}/2\mathbf{Z} harmonic spinors also appear in [ 12 ] ’s work on adiabatic limits of coassociative Kovalev–Lefschetz fibrations of G_{2} –manifolds [ 12 ] and [ 18 ] ’s work on branched double covers of special Lagrangian submanifolds [ 18 ] . In light of this, it is important to understand the universal moduli space of \mathbf{Z}/2\mathbf{Z} harmonic spinors (allowing for g , \gamma , and \nabla to vary). The fundamental issue is that D is only left semi-Fredholm (under mild assumptions; see 2.1 ), but not Fredholm—except in edge cases, e.g., if Z=\varnothing or n=2 . The naive expectation is that the \infty –dimensional cokernel of D can be compensated by wiggling the branching locus Z . In his PhD thesis, [ 34 , 33 ] has made some initial progress in this direction. [ 13 ] and [ 27 ] have developed a (partial) deformation theory for \mathbf{Z}/2\mathbf{Z} harmonic 1 –forms and \mathbf{Z}/2\mathbf{Z} harmonic spinors on spin 3 –manifolds respectively. There is work in progress by He, Parker and Walpuski to address this problem a bit more systematically. The present article should be considered infrastructure for this project [ 17 ] (and, hopefully, other applications as well). Here is a summary of the results contained in this article. Section 2 considers D as an unbounded operator D_{\mathrm{min}} on L^{2}\Gamma\lparen X{\backslash}Z,S\otimes{\mathfrak{l}}\rparen , the minimal extension , and systematically studies its closed extensions. The adjoint D_{\mathrm{max}}\coloneq D_{\mathrm{min}}^{*} is the maximal extension of D_{\mathrm{min}} . The closed extensions D_{R} of D_{\mathrm{min}} are classified by residue conditions , that is: closed subspaces R\subset\check{\mathbf{H}} of the Gelfand–Robbin quotient \check{\mathbf{H}}\coloneq\frac{\mathrm{dom}(D_{\mathrm{max}})}{\mathrm{dom}(D_{\mathrm{min}})}. Moreover, \check{\mathbf{H}} is equipped with a symplectic form G , the Green’s form , which controls the formation of adjoints. Within this framework it is also possible to describe which extensions D_{R} are Fredholm. The entire discussion only relies on D_{\mathrm{min}} being closed, densely defined, and symmetric as well as left semi-Fredholm. It is confined to the realm of abstract functional analysis and its purpose is to separate what is true for formal reasons from what is true for geometric reasons. Most of the observations in Section 2 can be found in [ 26 , 7 , 32 , 10 ] in some shape or form. Assuming that Z\subset X is a closed (cooriented) submanifold of codimension two, Section 3 constructs an isomorphism of symplectic Hilbert spaces \res\mskip 0.5mu\colon\thinspace\lparen\check{\mathbf{H}},G\rparen\cong\lparen\check{H}\Gamma\lparen Z,\check{S}\rparen,\check{\Omega}\rparen, the residue map , between the Gelfand–Robbin quotient and a Hilbert space of sections of a symplectic vector bundle over Z . The residue map extracts the leading order behavior of \phi\in\mathrm{dom}(D_{\mathrm{max}}) which is shown to be (at worst) comparable to \bar{z}^{-1/2} transversely to Z . With the help of the above it is possible to define spectral residue conditions, analogous to the APS boundary condition [ 3 ] , as well as local residue conditions. As a by-product this yields a variant of the bordism theorem, whose significance remains somewhat mysterious to the authors. Evidently, the construction in Section 3 is inspired by [ 4 ] ’s magnificent article [ 4 ] on boundary value problems for Dirac operators. Section 4 develops an L^{2} regularity theory on the scale of adapted Sobolev spaces \lparen H_{a}^{k+1}\Gamma\lparen X{\backslash}Z,S\otimes{\mathfrak{l}}\rparen\rparen_{k\in{\mathbf{N}}_{0}} . This scale is defined via conormal differential operators and the Dirac operator D . It gives rise to a graded Fréchet space H_{a}^{\infty}\Gamma\lparen X{\backslash}Z,S\otimes{\mathfrak{l}}\rparen which is tame in the sense of [ 15 , Part II Definition 1.3.2] —a prerequisite for using Nash–Moser theory. It is proved that if a residue condition R\subset\check{H}\Gamma\lparen Z,\check{S}\rparen is \infty –regular , then the extension D_{R} satisfies a variant of elliptic regularity together with elliptic estimates. For local residue conditions R , \infty –regularity can be verified using a symbolic criterion. In particular, this criterion applies to the Lagrangian local residue condition which is secretly at the heart of [ 34 , 27 ] . It is quite plausible that these results can be cobbled together using the powerful machines developed by [ 25 , 24 , 1 , 2 ] . However, the arguments in Section 4 are almost elementary and there should be some value in that. It should be possible, with suitable modifications, to extend the work in the present article to higher rank ramified Euclidean local systems; in particular: to flat Hermitian line bundles. In fact, Ammann–Große have work in progress in this direction and some instances of this appear in [ 29 ] ’s work on magnetic links [ 29 , 28 , 30 ] . Acknowledgements. The authors thank Siqi He and Greg Parker for helpful discussions, Jacek Rzemieniecki and Thibault Langlais for their careful proofreading of earlier versions of this article, and Bernd Ammann for discussions regarding his joint