Topological laminations on surfaces (Q2334431): Difference between revisions

Geodesic laminations on hyperbolic surfaces were introduced by William Thurston in the early seventies to provide a conclusion to the study of the space of simple closed curves. A geodesic lamination on a hyperbolic surface is a non-empty closed subset which is a union of simple pairwise disjoint complete geodesics. Since their introduction laminations occur in many contexts and have played central roles in low-dimensional topology. So far most of their use and analysis has been focused on measured geodesic laminations, geodesic laminations endowed with the additional structure of a transverse measure. However, there are results where geodesic laminations occur without a preferred transverse measure. Behind these results is the fact that geodesic laminations can be topologically characterized: A topological lamination \(\mathcal{T}\) on a surface \(\Sigma\) is a non-empty subset of \(\Sigma\) which is a union of simple and pairwise disjoint curves, all closed or two-way infinite, such that the curves are pairwise non-homotopic, they are all carried by a common train track \(\Gamma\), and \(\mathcal{T}\) is maximal for this \(\Gamma\) and remains so after any isotopies applied to the curves. The aim of this paper is to present a detailed exposition of the proof of the equivalence up to isotopy between the definition of a geodesic lamination and the definition of a topological lamination given above. The author also proves that a topological lamination carried by a train track \(\Gamma\) can be moved by an ambient isotopy into a regular neighborhood of \(\Gamma\). This result says that laminations can be described with respect to ribbon graphs relaxing their dependency to the surfaces themselves, and thus making laminations even more combinatorial objects.