Parametrization of sing \(\Theta\) for a Fano 3-fold of genus 7 by moduli of vector (Q2477151)

Let \(\Sigma \) be the spinor tenfold of degree 12 in \({\mathbb P}^{15}\), i.e., the unique closed orbit of Spin(10) in its projectivized half-spinor representation. The so called fundamental form allows one to identify \(\Sigma \) to its dual variety \({\Sigma}^{\vee}\subset {\mathbb P}^{15\vee}\). \textit{S. Mukai} [in: Algebraic geometry and commutative algebra. In honor of Masayoshi Nagata, Vol.I, Kinokuniya, Tokyo, 357--377 (1988; Zbl 0701.14044)] showed that a nonsingular section of \(\Sigma \) by a linear subspace \({\mathbb P}^m\subset {\mathbb P}^{15}\), \(m=6,\;7\), resp., 8, is a canonical curve, a \(K3\) surface, resp., a prime Fano threefold and that, conversely, a generic canonical curve of genus 7, a generic \(K3\) surface of degree 12 and, resp., any nonsingular prime Fano threefold of degree 12 (i.e., \(\text{Pic}X={\mathbb Z}(-K_X)\) and \((-K_X)^3=12\)) can be obtained as such a linear section. Now, let \(X={\mathbb P}^8\cap \Sigma \) be a prime Fano threefold of degree 12. The orthogonal linear section \(\Gamma := {\mathbb P}^{8\vee}\cap {\Sigma}^{\vee}\) is a canonical curve of genus 7 and the intermediate Jacobian \(\text{J}(X)\) is isomorphic to the Jacobian of \(\Gamma \). In their previous paper [Adv. Geom. 4, No. 3, 287--318 (2004; Zbl 1074.14039)], \textit{A. Iliev} and \textit{D. Markushevich} proved that, for a generic \(X\), the moduli space \({\text{M}}_X(2;1,5)\) of stable rank 2 vector bundles on \(X\) with \(c_1=-K_X\) and \(\text{deg}c_2=5\) is isomorphic to \(\Gamma \). \textit{A. Kuznetsov} [Math. Notes 78, No. 4, 537--550 (2005; Zbl 1111.14038)] remarked that this moduli space is fine and provided a natural universal bundle on it. In the paper under review, the authors show that, for a generic \(X\), the moduli space \({\text{M}}_X(2;1,6)\) is birational to the Brill-Noether locus \({\text{W}}^1_6(\Gamma )\) which is nothing else but the singular locus of the theta divisor \(\Theta \subset {\text{Pic}}^6\Gamma \) and that \({\text{M}}_X(2;1,6)\) is irreducible. In order to prove this result, they show that any \(E\in {\text{M}}_X(2;1,6)\) is generated by global sections at the generic point of \(X\) and that the zero locus of a non-zero global section of \(E\) lies in the closure of the family of elliptic sextics on \(X\). In order to handle the later family, they prove some auxiliary results about families of lower degree curves on \(X\). For example, they identify the curve of lines in \(X\) to \({\text{W}}^1_5(\Gamma )\) and show that the surface of conics in \(X\) is isomorphic to the symmetric power \({\Gamma}^{(2)}\). The proof of the irreducibility of the family of elliptic sextics in \(X\) uses a result of \textit{N. Perrin} [Courbes elliptiques sur la variété spinorielle de dimension 10, electronic preprint, \url{arXiv:math.AG/0409125}] on the irreducibility of the family of elliptic curves of fixed degree on \(\Sigma \). A technical tool used extensively throughout the paper is the Iskovskikh-Prokhorov-Takeuchi birational transformation \(X\dashrightarrow Q_3\subset {\mathbb P}^4\) that can be obtained by a blow-up with center a conic \(q\), followed by a flop and a contraction of one divisor onto a curve \({\Gamma}_q\subset Q_3\) isomorphic to \(\Gamma \). Using this transformation, one can, for example, associate to any degree \(d\) curve \(C\) in \(X\) its Abel-Jacobi image \(\text{AJ}(C)\in {\text{Pic}}^{5d} \Gamma \).