Torsion and critical metrics on contact three-manifolds

General geometric structures on manifolds (almost complex, almost product structures, etc.) (53C15) Special Riemannian manifolds (Einstein, Sasakian, etc.) (53C25) Critical metrics (58E11)

Review text

Let M be a 3-dimensional compact contact manifold with contact form \(\omega\) and characteristic vector field \(X_ 0\). Let \({\mathcal M}\) denote the set of all associated metrics to \(\omega\). For any \(g\in {\mathcal M}\) let r denote the scalar curvature, S the Ricci tensor, \(\tau\) the Lie derivative of g with respect to \(x_ 0\) and \(\sigma\) the restriction of \(S(X_ 0,)\) to the contact subbundle. The first result of the paper is that \(g\in {\mathcal M}\) is a critical point of \({\mathcal F}(g)=\int_{M}r dv\) considered as a function on \({\mathcal M}\) if and only if \(\nabla_{X_ 0}\tau =0\). For contact manifolds of general dimension, this problem was considered by \textit{A. J. Ledger} and the reviewer [J. Aust. Math. Soc., Ser. A 41, 404-410 (1986; Zbl 0611.53031)] and the critical point condition here can be obtained from the general one using the form of the curvature tensor in dimension 3. The second result of the paper is to give an alternate proof of the theorem of \textit{S. S. Chern} and \textit{R. S. Hamilton} [Lect. Notes Math. 1111, 279-305 (1985; Zbl 0561.53039)] that \(g\in {\mathcal M}\) is critical for \(\int_{M}W dv\) where W is the Webster scalar curvature if and only if \(X_ 0\) is Killing with respect to g, i.e. (\(\omega\),g) is K-contact. Remarks are also made concerning the functional \(\int_{M}| \tau |^ 2 dv\) which has been studied by both Chern and Hamilton and the reviewer in the above mentioned papers. Thirdly the author studies \(\omega\)-Einstein metrics, i.e. \(g\in {\mathcal M}\) for which \(S=ag+b\omega \otimes \omega.\) For M a 3-dimensional compact contact manifold \(g\in {\mathcal M}\) is \(\omega\)-Einstein if and only if g is critical for \({\mathcal F}\) and \(\sigma =0\). Several further results are given, the most interesting of which is the following: Let M be a 3- dimensional compact contact manifold with K-contact structure (\(\omega\),g). If any one of the following holds: a) \(W>0\), b) \(r>-2\), c) \(S+2g\) positive definite, d) the f-sectional curvature \(>-3,\) then M admits a K-contact metric structure \(({\tilde\omega},\tilde g)\) of positive curvature. Thus if M is simply connected it is diffeomorphic to the 3-sphere.

Recommendations

Cites work

Cited in

(25)

This page was built for publication: Torsion and critical metrics on contact three-manifolds

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q921611)