The height invariant of a four-parameter semitoric system with two focus-focus singularities (Q2022531)

Let \((M,\omega)\) be a connected symplectic 4-manifold. An integrable system \(F:=(L,H):M\to \mathbb{R}^2\) on \(M\) is called a \textit{semitoric system} if \(L\) is a proper map inducing an effective circle action, and the singularities of \(F\) are non-degenerate with no hyperbolic component. It was established in [\textit{A. Pelayo} and \textit{S. Vũ Ngọc}, Invent. Math. 177, No. 3, 571--597 (2009; Zbl 1215.53071)] that semitoric systems are completely classified in terms of five invariants. Their computation is in general complicated, especially when one wants to obtain these invariants simultaneously for a family of semitoric systems depending on several parameters. In the present paper, the authors compute three of the five invariants for a family of semitoric systems on \(M=\mathbb{S}^2\times \mathbb{S}^2\) endowed with the symplectic form \(\omega=-(R_1 \omega_{\mathbb{S}^2}\oplus R_2 \omega_{\mathbb{S}^2})\), which depends on two geometric parameters (\(R_1,R_2>0\) distinct) and two coupling parameters \((s_1,s_2\in [0,1])\). The functions defining the family are explicitly given by \[ \left\{ \begin{array}{rl} L(m):=&R_1 z_1+R_2 z_2 \,, \\ H(m):=&(1-2s_1)(1-s_2)z_1+(1-2s_1)s_2 z_2 \\ &+ 2(s_1+s_2-s_1^2-s_2^2)(x_1x_2+y_1y_2)\,, \end{array} \right. \] where \(m=(x_1,y_1,z_1,x_2,y_2,z_2)\in M\) for cartesian coordinates on \(\mathbb{S}^2\subset \mathbb{R}^2\). (This defines a semitoric system for a generic choice of coupling parameters \(s_1,s_2\), see Theorem 15.) The computation of the first invariant (number of focus-focus singularities) is done in subsection 3.1 and builds on results from [\textit{S. Hohloch} and \textit{J. Palmer}, J. Geom. Mech. 10, No. 3, 331--357 (2018; Zbl 1428.37053)]; the second invariant (polygon invariant) is obtained in Theorem 21 and follows from [\textit{Y. Le Floch} and \textit{Á. Pelayo}, J. Nonlinear Sci. 29, No. 2, 655--708 (2019; Zbl 1417.53085)]. The explicit derivation of the third invariant (the height invariant) is given in Theorem 22. This computation of the height invariant in the presence of two focus-focus points and four parameters is the first of its kind to have appeared.