Fischer-Marsden conjecture on real hypersurfaces in the complex quadric (Q6134668)

Let \(M\) be an \(n\)-dimensional orientable compact manifold and denote by \(\mathfrak g\) the set of all Riemannian metrics \(g\) which preserve the volume. For every \(g\in\mathfrak g\), let \(\mathcal S_g\) and \(dv_g\) respectively denote the corresponding scalar curvature and the volume element. A critical point of the total scalar curvature functional \[ g\mapsto \int_g\mathcal S_g \, dv_g \] is an Einstein metric. The linearization of the above scalar curvature functional is given by \[ \mathcal S'_g(h)=-\Delta_g(\mathrm{tr}_gh)+\mathrm{div}(\mathrm{div}(h))-g(h, \mathrm{Ric}_g) \] for \((0,2)\)-type symmetric bilinear tensors \(h\) on \(M\), where \(\mathrm{div}\) is the divergence, \(\Delta_g\) is the negative Laplacian of \(g\) and \(\mathrm{Ric}_g\) is the Ricci tensor of \(g\). It can be shown that the formal \(L^2\) adjoint of the operator \(\mathcal S'_g\) is of the form \[ {\mathcal S'_g}^{\ast}(\nu)=-(\Delta_g\nu)\,g-\nu\, \mathrm{Ric}_g+\mathrm{Hess}_g\nu \] for smooth functions \(\nu\) defined on \(M\), where \(\mathrm{Hess}_g\nu\) denotes the Hessian of \(\nu\). The equation \( {\mathcal S'_g}^{\ast}(\nu)=0\) is called Fischer-Marsden equation (note that the solutions of such equations are of the form \((g, \nu)\)). It is known that on a complete Riemannian manifold \(M\), if the Fischer-Marsden equation has no nontrivial solution, then the corresponding scalar curvature \(\mathcal S_g\) of the Riemannian metric \(g\) is constant. \textit{A. E. Fischer} and \textit{J. A. Wolf} conjectured in [Bull. Am. Math. Soc. 80, 92--97 (1974; Zbl 0303.53046)] that ``every compact Riemannian manifold which admits a non-trivial solution of \( {\mathcal S'_g}^{\ast}(\nu)=0\) is an Einstein manifold.'' This conjecture is studied widely in various cases in the literature. The present paper investigates it in the case of real hypersurfaces in the complex quadric \[ Q^m:=\mathrm{SO}_{m+2}/\mathrm{SO}_2\mathrm{SO}_m. \] There exist two main classes of such hypersurfaces. The first class comprises contact hypersurfaces which are tubes over a unit sphere in \(Q^m\). The second class consists of the hypersurfaces with isometric Reeb flow which equivalently are the tubes over \(\mathbb CP^k\) in \(Q^{2k}\). None of these subclasses contain Einstein manifolds and thus it is interesting to check the Fischer-Marsden conjecture for them. As a main result the author shows that the Fischer-Marsden equation does not admit any non-trivial solution for the contact hypersurfaces in \(Q^m\) with \(m\geq 3\) and for the real hypersurfaces with isometric Reeb flow in \(Q^m\) with \(m\geq 4\). Thus, in these cases the Fischer-Marsden conjecture is correct.