Unfolding of wild character varieties
Kazuki Hiroe, Daisuke Yamakawa · 2025
Read Unfolding of wild character varieties by Kazuki Hiroe, Daisuke Yamakawa (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-2511-01221.txt.
Introduction
Kazuki Hiroe and Daisuke Yamakawa study wild character varieties, spaces that parametrize certain equations with irregular singularities on compact Riemann surfaces. The tame cousins, with only regular singularities, are far better understood, and bridging the two has been an open problem in the geometry of these moduli spaces.
The authors construct Poisson maps from wild to tame character varieties by unfolding each irregular singularity into regular ones. They then show these unfolding maps are Poisson birational equivalences, meaning the wild and tame spaces match up at the level of their Poisson geometry away from small excluded sets. The result answers yes to a conjecture posed by Klimes, Paul, and Ramis.
Catalog record
| CREATOR | Kazuki Hiroe, Daisuke Yamakawa |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-01221
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Unfolding of wild character varieties Kazuki Hiroe, Daisuke Yamakawa Abstract . In this paper, we study wild character varieties on compact Riemann surfaces and construct Poisson maps from wild to tame character varieties by unfolding irregular singularities into regular ones. Furthermore, we show that these unfolding Poisson maps induce Poisson birational equivalences between wild and tame character varieties. This result provides an affirmative answer to a conjecture posed by Klimeš, Paul, and Ramis. 2020 Mathematics Subject Classification Primary 14M35; Secondary 34M40, 14J42, 32G34, 53D30 1. Introduction This article investigates a geometric aspect of the confluence of singularities in meromorphic connections, formulated through wild character varieties. The main motivation is to construct Poisson and birational correspondences between wild character varieties and the character varieties arising from the unfolding of irregular singularities. We now outline the main results of this article. Let G be a complex reductive group with a fixed maximal torus T . Define \mathbf{\Sigma}=(\Sigma,\mathbf{a};{}^{1}Q,\ldots,{}^{m}Q) as an untwisted irregular curve in the sense of Boalch [ 5 ] , consisting of a compact Riemann surface \Sigma of genus g , a finite set of marked points \mathbf{a}=\{a_{1},\ldots,a_{m}\}\subset\Sigma , and a collection of irregular types {}^{i}Q . Fixing a local coordinate z_{i} centered at each marked point a_{i} , we write {}^{i}Q=\sum_{j=1}^{r_{j}}{}^{i}Q_{j}z^{-j}_{i}\in z_{i}^{-1}\mathfrak{t}[z_{i}^{-1}] and let \mathbf{H}=\prod_{i=1}^{m}Z_{G}(^{i}Q) denote the product of stabilizers of all coefficients {}^{i}Q_{j}\in\mathfrak{t} appearing in each {}^{i}Q . The wild character variety \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}) associated with \mathbf{\Sigma} is then defined as a Poisson variety. Furthermore, given a conjugacy class \mathcal{C}=\prod_{i=1}^{m}\mathcal{C}_{i}\subset\mathbf{H} , we define the symplectic wild character variety \mathcal{M}_{\mathrm{B}}^{{\mathrm{st}}}(\mathbf{\Sigma},\mathcal{C}) as an open subset of the closed Poisson subvariety \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma},\overline{\mathcal{C}}) of \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}) , corresponding to the closure \overline{\mathcal{C}} of \mathcal{C} . For precise definitions, see Section 4 . Let us consider the unfolding of the wild character varieties. Let \mathbf{\Sigma}^{\prime}=(\Sigma,\mathbf{b};\mathbf{0}) be the irregular curve with \sum_{i=1}^{m}(r_{j}+1)=|\mathbf{b}| marked points, where all irregular types are trivial. In other words, \mathbf{\Sigma}^{\prime} is simply a Riemann surface with marked points. The associated wild character variety \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime}) is then a standard (tame) character variety that parametrizes the isomorphism classes of semisimple G -representations of the fundamental group \pi_{1}(\Sigma\backslash\mathbf{b}) . Next, we unfold the conjugacy class \mathcal{C} as follows. For each i=1,\ldots,m and j=1,\ldots,r_{i} , choose {}^{i}t_{j}\in T such that Z_{G}(^{i}t_{j})=Z_{G}({}^{i}Q_{j},{}^{i}Q_{j+1},\dots,{}^{i}Q_{r_{i}}),\quad Z_{G}(h({}^{i}t_{1}{}^{i}t_{2}\cdots{}^{i}t_{r_{i}})^{-1})\subset Z_{G}(^{i}Q)\ (h\in\mathcal{C}_{i}). See Appendix A for the existence of such elements. Let {}^{i}\mathfrak{C}_{j} be the G -conjugacy class containing {}^{i}t_{j} , and let {}^{i}\mathfrak{C}_{0} be the G -conjugacy class containing \mathcal{C}_{i}({}^{i}t_{1}{}^{i}t_{2}\cdots{}^{i}t_{r_{i}})^{-1} . We then obtain the character varieties \mathcal{M}_{\mathrm{B}}^{{\mathrm{st}}}(\mathbf{\Sigma}^{\prime},\mathfrak{C}) and \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime},\overline{\mathfrak{C}}) associated with the conjugacy class \mathfrak{C}=\prod_{i,j}{}^{i}\mathfrak{C}_{j} . The main results of this article are as follows. As a consequence of Theorem 8.1 , we relate the wild character variety \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma},\overline{\mathcal{C}}) and the unfolded character variety \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime},\overline{\mathfrak{C}}) in a Poisson sense, namely, we obtain a Poisson map \Upsilon_{({}^{i}t_{j})}\colon\mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma},\overline{\mathcal{C}})\rightarrow\mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime},\overline{\mathfrak{C}}). Furthermore, Corollary 8.2 shows that the following Poisson birational equivalence holds: Theorem 1.1 . Suppose that \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma},\overline{\mathcal{C}}) , \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime},\overline{\mathfrak{C}}) are both irreducible and \mathcal{M}_{\mathrm{B}}^{\mathrm{st}}(\mathbf{\Sigma},\mathcal{C}) , \mathcal{M}_{\mathrm{B}}^{\mathrm{st}}(\mathbf{\Sigma}^{\prime},\mathfrak{C}) are both nonempty. Then \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma},\overline{\mathcal{C}}) and \mathcal{M}_{\mathrm{B}}(\mathbf{\Sigma}^{\prime},\overline{\mathfrak{C}}) are Poisson birationally equivalent. Namely, there exist nonempty open subsets of them which are isomorphic as Poisson varieties. We now provide some background and related work. In [ 11 ] , a deformation of moduli spaces of meromorphic connections on the trivial G -bundle over the Riemann sphere \mathbb{P}^{1} was constructed via the unfolding of irregular singularities. It was subsequently shown that any moduli space of meromorphic G -connections with unramified (untwisted) irregular singularities admits a deformation to the moduli space of Fuchsian G -connections. Our main theorem gives a generalization of this result in the context of wild character varieties and further establishes Poisson birational equivalences among the moduli spaces arising from such deformations. In related work, Klimeš [ 12 ] showed the existence of a birational transformation between character varieties associated with the Painlevé V and Painlevé VI equations in the context of nonlinear Stokes phenomena for Painlevé equations. Building on this, Paul and Ramis [ 14 ] showed that the birational transformation by Klimeš is in fact symplectic, and they posed an open problem suggesting that character varieties for other types of Painlevé equations should admit similar symplectic and birational maps arising from the unfolding of their irregular singularities. Our main theorem provides an answer to this problem in the case of untwisted wild character varieties. Throughout this article, we fix a complex reductive group G together with an \operatorname{Ad} -invariant non-degenerate symmetric bilinear form (\bm{\cdot},\bm{\cdot}) on its Lie algebra \mathfrak{g}=\operatorname{Lie}G . A variety means a (possibly reducible) complex algebraic variety. The identity component of an algebraic group H is denoted by H^{0} . 2. Quasi-Hamiltonian geometry In this section we briefly recall some basic notions and facts in (algebraic) quasi-Hamiltonian geometry; for more details, see [ 1 , 2 , 8 , 5 , 13 ] . 2.1. Quasi-Poisson and quasi-Hamiltonian structures For a (possibly singular) G -variety M with tangent sheaf \Theta_{M}=\operatorname{\mathcal{H}\hskip-1.00006pt\mathit{om}}(\Omega^{1}_{M},\mathcal{O}_{M}) , we denote by \mathfrak{g}\to\Gamma(M,\Theta_{M}) , \xi\mapsto\xi_{M} the corresponding infinitesimal action; for \xi\in\mathfrak{g} , the vector field \xi_{M} is characterized by \xi_{M}(df)(p)=\left.\frac{d}{dt}f(e^{-t\xi}\cdot p)\right|_{t=0}\quad(p\in M,\ f\in\mathcal{O}_{M,p}). For each k\in\mathbb{Z}_{>0} it induces a map (\bm{\cdot})_{M}\colon\bigwedge^{k}\mathfrak{g}\to\Gamma(M,\bigwedge^{k}\Theta_{M}) . Define a homomorphism of \mathcal{O}_{M} -modules (\bm{\cdot})_{M}\colon\mathcal{O}_{M}\otimes_{\mathbb{C}}\bigwedge^{k}\mathfrak{g}\to\bigwedge^{k}\Theta_{M} by f\otimes\xi\mapsto f\xi_{M} . Let (\bm{\cdot})^{\vee}\colon\mathfrak{g}^{*}\to\mathfrak{g} , \alpha\mapsto\alpha^{\vee} be the inverse of the isomorphism \xi\mapsto(\xi,\bm{\cdot}) . Define \mathfrak{g} -valued vector fields \delta^{L},\delta^{R}\in\Gamma(G,\Theta_{G}\otimes_{\mathbb{C}}\mathfrak{g}) on G by \delta^{L}_{x}(\theta)=(L_{x}^{*}\theta)^{\vee},\quad\delta^{R}_{x}(\theta)=(R_{x}^{*}\theta)^{\vee}\quad(x\in G,\ \theta\in T_{x}^{*}G). Put \delta=\tfrac{1}{2}(\delta^{L}+\delta^{R}) . Definition 2.1 . (1) A quasi-Poisson G -variety is a G -variety M equipped with a G -invariant skew-symmetric bracket operation \{\bm{\cdot},\bm{\cdot}\}\colon\mathcal{O}_{M}\otimes_{\mathbb{C}}\mathcal{O}_{M}\to\mathcal{O}_{M}, on the structure sheaf \mathcal{O}_{M} satisfying (1) \displaystyle\{f,gh\} \displaystyle=g\{f,h\}+\{f,g\}h, (2) \displaystyle\{f,\{g,h\}\}+\{g,\{h,f\}\}+\{h,\{f,g\}\} \displaystyle=\chi_{M}(df,dg,dh)\qquad(f,g,h\in\mathcal{O}_{M}), where \chi\in\bigwedge^{3}\mathfrak{g} is defined by \chi(\alpha,\beta,\gamma)=\langle\alpha,[\beta^{\vee},\gamma^{\vee}]\rangle\quad(\alpha,\beta,\gamma\in\mathfrak{g}). The bracket \{\bm{\cdot},\bm{\cdot}\} is called a quasi-Poisson bracket . (2) For a quasi-Poisson G -variety M , a G -equivariant morphism \mu\colon M\to G (where G acts on itself by conjugation) is called a moment map if the bivector field \Pi\in\Gamma(M,\bigwedge^{2}\Theta_{M}) corresponding to the bracket \{\bm{\cdot},\bm{\cdot}\} satisfies \Pi(\mu^{*}\theta,\bm{\cdot})=\left(\delta(\theta)\circ\mu\right)_{M}\quad(\theta\in\Omega^{1}_{M}). A quasi-Poisson G -variety equipped with a moment map is called a Hamiltonian quasi-Poisson G -variety . Note that if G is abelian (e.g. G=\{1\} ), then \chi=0 and hence a quasi-Poisson structure is a usual Poisson structure. Just as a non-degenerate Poisson structure comes from a symplectic structure, so some class of Hamiltonian quasi-Poisson structures come from the following geometric objects. Definition 2.2 . A quasi-Hamiltonian G -space is a smooth G -variety equipped with a G -invariant two-form \omega on M and a G -equivariant morphism \mu\colon M\to G (where G acts on itself by conjugation) satisfying the following conditions. (QH1) d\omega=\tfrac{1}{12}(\mu^{-1}d\mu,[\mu^{-1}d\mu,\mu^{-1}d\mu]) . (QH2) \omega(\xi_{M},\bm{\cdot})=(\mu^{-1}d\mu+d\mu\,\mu^{-1},\xi) for all \xi\in\mathfrak{g} . (QH3) \operatorname{Ker}\omega_{p}\cap\operatorname{Ker}(d\mu)_{p}=\{0\} for any p\in M . The two-form \omega is called the quasi-Hamiltonian two-form and \mu is called the moment map . If G is abelian, then the three axioms imply that \omega is a symplectic form. Theorem 2.3 ( [ 1 , Theorem 10.3] , [ 9 , Corollary 3.22] ) . For any quasi-Hamiltonian G -space (M,\omega,\mu) , there exists a unique quasi-Poisson G -structure \{\bm{\cdot},\bm{\cdot}\} on M with \mu a moment map such that the corresponding bivector field \Pi\in\Gamma(M,\bigwedge^{2}\Theta_{M}) satisfies \Pi(\iota_{v}\omega,\bm{\cdot})=v-\frac{1}{4}\left((\mu^{-1}d\mu-d\mu\,\mu^{-1})(v)\right)_{M}\quad(v\in\Theta_{M}). A Hamiltonian quasi-Poisson structure coming from a quasi-Hamiltonian structure in the above way is said to be non-degenerate ; see [ 1 ] for an explicit characterization of non-degeneracy. Example