Two remarks on the asymptotic Hopf invariant (Q748962): Difference between revisions

A vector field U on a three-dimensional manifold with a volume form \(\mu\) is called homologous to zero if for a 1-form \(\alpha\) : \(i_ u\mu =d\alpha\). \(\int \alpha \wedge d\alpha\) is called the asymptotical Hopf invariant of the vector field u and \(L(u,v)=\int \alpha \wedge d\beta\) is called the asymptotical linking number of the vector fields u, v which are homologous to zero, where \(i_ u\mu =d\alpha\), \(i_ v\mu =d\beta\). The author notices that the Godbillon-Vey class of a codimension one foliation is the asymptotical Hopf invariant of a certain vector field associated to the foliation. He poses the question: Is there such a representation for highest characteristic classes? \textit{D. Bennequin} [Astérisque 107/108, 87-161 (1983; Zbl 0573.58022)] defined an invariant of a contractible curve which is transversal to a contact structure on a three-dimensional manifold. Now the author defines an asymptotical Bennequin invariant of vector fields \(\xi\), u (each of them homologous to zero) \[ \frac{d^ 2}{dt^ 2}(L(\phi_ t(\xi),\xi)- L(\xi,\xi)) \] where \(\phi_ t\) is the flow of the vector field u. Some properties of this invariant are established. At the end of the paper two open questions are formulated. The first one is: Let \(\xi\) be a vector field on a manifold \(M^{2n+1}\) and \(i_{\xi}\mu =(d\lambda)^ n\). Then the asymptotical Hopf invariant is defined: \(\int {\dot \lambda}\wedge d\lambda\). How to describe it geometrically?