Morse area and Scharlemann-Thompson width for hyperbolic 3-manifolds

From MaRDI portal
Publication:2634643



Abstract: Scharlemann and Thompson define a numerical complexity for a 3-manifold using handle decompositions of the manifold. We show that for compact hyperbolic 3-manifolds this is linearly related to a definition of metric complexity in terms of the areas of level sets of Morse functions.


Let \(M\) be a closed hyperbolic \(3\)-manifold. The area of a Morse function \(f: M\to{\mathbb R}\) is defined as the maximum area of level sets \(f^{-1}(t)\) over \(t\in{\mathbb R}\), and the Morse area of \(M\) is defined as the infimum of the area over all Morse functions. While a Morse function with only one maximum and minimum yields a Heegaard splitting of \(M\), a general Morse function does yield what the authors call a linear splitting: a decomposition of \(M\) into a linearly ordered sequence \(C_1,\dots,C_{2n}\) of compression bodies such that the upper boundary of \(C_{2i+1}\) equals the upper boundary of \(C_{2i+2}\) and the lower boundary of \(C_{2i+1}\) equals the lower boundary of \(C_{2i}\), and the lower boundaries of \(C_1\) and \(C_{2n}\) are empty. (Recall that a compression body \(C\) is built from its upper boundary \(\partial^+C\) by attaching some 2-handles to \(\partial^+C\times\left[0,1\right]\), and capping off all resulting spherical boundary components with \(3\)-balls.) The width of such a splitting is the sum of the genera of the upper boundary surfaces, and the linear width or Scharlemann-Thompson width of \(M\) is the minimal width over all linear splittings. The authors prove that there is a constant \(K>0\) such that \[ K(\mathrm{linear\;width}(M))\leq \mathrm{Morse\;area}(M) \] holds for all closed hyperbolic \(3\)-manifolds \(M\). Assuming an announced (but not yet fully proved) result of \textit{J. T. Pitts} and \textit{J. H. Rubinstein} [Proc. Cent. Math. Anal. Aust. Natl. Univ. 10, 163--176 (1986; Zbl 0602.49028)] they also obtain an upper bound \[ \mathrm{Morse\;area}(M)\leq 4\pi(\mathrm{linear\;width}(M)). \] The approach to the proof (for the first inequality) is to use a Voronoi decomposition of the thick part, to approximate level sets of \(f\) by surfaces which are unions of faces of Voronoi cells, and thus to get a linear splitting of the thick part for which one can control the genus of the boundary surfaces in the terms of the area of level sets. For the thin part, its topology is known by the Margulis lemma and the complexity of its linear splittings is controlled using work of Kobayashi-Rieck [\textit{T. Kobayashi} and \textit{Y. Rieck}, ``A linear bound on the tetrahedral number of manifolds of bounded volume (after Jørgensen and Thurston), Contemporary Mathematics 560, 27--42 (2011; Zbl 1335.57028)].











This page was built for publication: Morse area and Scharlemann-Thompson width for hyperbolic 3-manifolds

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2634643)