Adiabatic limits and Kazdan-Warner equations (Q1800867)

The vortex equations on Riemann surfaces originate from the Ginzburg-Landau model of superconductivity. They were first studied by \textit{A. Jaffe} and \textit{C. Taubes} [Vortices and monopoles. Structure of static gauge theories. Boston etc.: Birkhäuser (1980; Zbl 0457.53034)]; since then there has been a considerable body of work aimed at understanding these equations and their many generalizations. The goal of this article (stemming from the PhD thesis of the author) is to study one such generalization in the context of adiabatic limit and compactification in two and three dimensional gauge theories. In particular, let \(\Sigma\) be a closed Riemann surface, and \(L \to \Sigma\) an Hermitian line bundle. Then a classical vortex is a pair \((A,\varphi)\) where \(A\) is a connection on \(L\) and \(\varphi \in \Gamma (L)\) a section of \(L \to \Sigma\), satisfying the following equations, where \(F_A\) is the curvature form of \(A\) and the Hodge dual of this is denoted as \(\star F_A\): \[ \overline{\partial}_A \varphi = 0, \quad i (\star F_A) = 1 - |\varphi |^2 . \] The present paper studies a generalization of these. Let \(E_1 , \dots , E_N\) be auxiliary bundles over \(\Sigma\), \(B_1 , \dots , B_N\) connections over these, \(k_1 , \dots , k_N\) integers. Now \[ \varphi = (\varphi^1 , \dots , \varphi^N) \in \Gamma ( \bigoplus_{j=1}^N E_j \otimes L^{\otimes k_j} ) . \] Moreover there are real parameters \(\tau\) and \(\varepsilon >0\). The equations under study, which fit within the general setting of framed vortex equations, are then \[ {\overline{\partial}}_{A \otimes B_j} \varphi^j = 0 \;(j = 1,\dots,N) , \quad \varepsilon^2 \, i (\star F_A) + \sum_{j=1}^N k_j \, |\varphi^j |^2 + \tau = 0 . \tag{\(*\)} \] (For \(N=k_1=\varepsilon= - \tau = 1\) one recovers the classical vortex equations.) A necessary condition for these to admit a solution is that \(\{ k_1 , \dots , k_N ; \tau \}\) have mixed signs. The main result of the paper is the following Theorem. Let \((A_i , \varphi_i , \varepsilon_i )\) be a sequence of solutions to \((*)\) such that \(\varepsilon_i \to 0\) and the sequence of \(L^2\) norms \(\| \varphi_i \|\) is bounded. Then there is a finite set of points \(D \subset \Sigma\) such that after passing to a subsequence and applying gauge transformations, \((A_i , \varphi_i )\) converges in \(C^\infty_{\mathrm{loc}}\) on \(\Sigma \backslash D\). The limit \((A,\varphi)\) satisfies \((*)\) with \(\varepsilon = 0\) on \(\Sigma \backslash D\). This theorem also allows to tackle and solve a number of questions in Seiberg-Witten theory (in this context, and for the relations of generalized vortices with Seiberg-Witten theory, see also [the author, ``Seiberg-Witten monopoles with multiple spinors on a surface times a circle'', Preprint, \url{arXiv:1701.07942}; the author and \textit{T. Walpuski}, ``On the existence of harmonic \(\mathbb Z_2\) spinors'', Preprint, \url{arXiv:1710.06781}]), and leads to the first known example of a non-compactness phenomenon predicted by \textit{A. Haydys} and \textit{T. Walpuski} [Geom. Funct. Anal. 25, No. 6, 1799--1821 (2015; Zbl 1334.53039)].