Birational geometry of moduli spaces of stable objects on Enriques surfaces (Q2297380)

The moduli space of stable sheaves on a smooth projective surface is an interesting geometric object and has been studied for a long time. By using the notion of Bridgeland stability condition on triangulate category and its wall-crossing behaviour [\textit{T. Bridgeland}, Ann. Math. (2) 166, No. 2, 317--345 (2007; Zbl 1137.18008)], there are lots of progress in this direction, including the birational geometry of the moduli space when the surface is \(K3\) [\textit{A. Bayer} and \textit{E. Macrì}, Invent. Math. 198, No. 3, 505--590 (2014; Zbl 1308.14011)] or \(\mathbb{P}^2\) [\textit{C. Li, X. Zhao}, Geom. Topol. 23, No. 1, 347--426 (2019; Zbl 1456.14016)]. The paper under review continues the idea for the \textit{generic} Enriques surface. Let \(Y\) be an Enriques surface and \(\pi: \widetilde{Y} \rightarrow Y\) be the universal covering map by the \(K3\) surface \(\widetilde{Y}\). Assume that \(Y\) is generic, that is \(\mathrm{Pic}(\widetilde{Y})=\pi^*\mathrm{Pic}(Y)\). Let \(v\) be a Mukai vector so that its pullback \(\pi^*(v)\) is primitive. The main Theorem, Theorem 4.5, established the birational equivalence of two moduli spaces \(M^Y_\sigma(v)\) and \(M^Y_\tau(v)\) for two generic stability conditions \(\sigma, \tau \in \mathrm{Stab}^{\dag}(Y) \) with respect to the Mukai vector \(v\). To prove Theorem 4.5, the author uses two main ideas. The first idea is to use the notion of constant cycle subvariety. A subvariety is called a constant cycle if all its points become rationally equivalent in the ambient variety. By using the result of [\textit{A. Marian, X. Zhao}, Épijournal de Géom. Algébr., EPIGA Journal Profile 4, Article No. 3, 5 p. (2020; Zbl 1442.14035)], the author shows that the image of the morphism \(\pi^*: M^Y_\sigma(v) \rightarrow M^{\widetilde{Y}}_{\widetilde{\sigma}}(\pi^*(v))\) is a constant cycle Lagrangian. Here \(\widetilde{\sigma}\) is the induced Bridgeland stability on \(\widetilde{Y}\) [\textit{E. Macrì} et al., J. Algebr. Geom. 18, No. 4, 605--649 (2009; Zbl 1175.14010)]. The second idea is to show the corresponding birational morphism \(f: M^{\widetilde{Y}}_{\widetilde{\sigma}_+}(\pi^*(v)) \dashrightarrow M^{\widetilde{Y}}_{\widetilde{\sigma}_-}(\pi^*(v))\) on the \(K3\) surface \(\widetilde{Y}\) is \(i^*\)-equivariant. Here \(i^*\in \mathrm{Aut}(\mathrm{D}^{\mathrm{b}}(\widetilde{Y}))\) is the induced involution by the map \(\pi\). The assumption that \(Y\) is generic is used in the proof of \(i^*\)-equivariance. The moduli spaces \(M^Y_{\sigma_\pm}(v)\) can be identified as fixed set of the involution \(i^*\). Since \(f\) is \(i^*\)-equivariant, then its restriction to the constant cycle Lagrangian \(f|_{M^Y_{\sigma_+}(v)}: M^Y_{\sigma_+}(v) \dashrightarrow M^Y_{\sigma_-}(v)\) gives the birational morphism. As an application of the main Theorem, the author shows (in Theorem 4.7) that for an arbitrary Enriques surface \(Y\) and a primitive Mukai vector \(v\) of odd rank, then \(M^Y_\sigma(v)\) is birational to some Hilbert scheme of points \(\mathrm{Hilb}^n(Y)\). As another application, the author shows (in Lemma 4.11) that the existence of global Bayer-Macrì map \(\ell: \mathrm{Stab^{\dag}(Y)} \rightarrow \mathrm{NS}(M_\sigma(v))\) [\textit{A. Bayer} and \textit{E. Macrì}, Invent. Math. 198, No. 3, 505--590 (2014; Zbl 1308.14011); \textit{W. Liu}, Kyoto J. Math. 58, No. 3, 595--621 (2018; Zbl 1412.14009)]. Moreover, the author shows the nef and semiample divisors \(\ell_{\sigma_0,\pm}\in \mathrm{NS}(M_{\sigma_{\pm}}(v)\) are big. \textit{H. Nuer} and \textit{K. Yoshioka} [Adv. Math. 372, Article ID 107283, 118 p. (2020; Zbl 1454.14041)] obtained more general results of birational equivalence of \(M^Y_\sigma(v)\) and \(M^Y_\tau(v)\) by a different method without assumptions that \(Y\) is generic and \(v\) is primitive.