Homological shadows of attracting laminations (Q1784004)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    Homological shadows of attracting laminations
    scientific article

      Statements

      Homological shadows of attracting laminations (English)
      0 references
      0 references
      21 September 2018
      0 references
      For a free group \(F_n\) and a fully irreducible automorphism \(f\in\mathrm{Aut}(F_n)\) whose action on \(H_1(F_n;\mathbb{Z})\) has finite order, the author introduces a homological version of the attracting lamination of \(f\), together with a measure supported at a single point. More precisely, for any \(x\in F_n\) with infinite \(f\)-orbit and any \(k\geq \mathbb{Z}_+\), the author takes the reduced word \(f^k(x)\in F_n\), and its induced (constant speed) path \(\tilde{S}f^k(x):[0,1]\to H_1(F_n;\mathbb{R})\cong \mathbb{R}^n\) in homology. Let \(\tilde{S}_kf^k(x)\) be the the \(\frac{1}{k}\)-scaling of \(\tilde{S}f^k(x)\), and let \(S_kf^k(x)\) be the image of \(\tilde{S}_kf^k(x)\). Then the author proves that there exists a convex polytope \(S_{\infty}f\subset H_1(F_n;\mathbb{R})\) with rational vertices, such that \[ \lim_{k\to \infty}S_kf^k(x)=S_{\infty}f \] in the Hausdorff topology. Moreover, for the standard Lebesgue measure \(\lambda\) on \([0,1]\), its push forward by \(\tilde{S}_kf^k(x)\) provides a measure \(D_kf^k(x)=(\tilde{S}_kf^k(x))_*\lambda\) on \(H_1(F_n;\mathbb{R})\). Then the author proves that there exists a probability measure \(D_{\infty}f\) on \(H_1(F_n;\mathbb{R})=\mathbb{R}^n\) supported at a point in \(\mathbb{Q}[\rho]^n\), such that \[ \lim_{k\to \infty}D_kf^k(x)=D_{\infty}f \] in the weak-\(*\) topology. Here \(\rho\in \mathbb{R}_+\) is the dilatation of \(f\). The pair \((S_{\infty}f, D_{\infty}f)\) is a homological version of \textit{M. Bestvina} et al.'s work [Geom. Funct. Anal. 7, No. 2, 215--244 (1997; Zbl 0884.57002)] on attracting laminations of fully irreducible automorphisms on free groups. So \((S_{\infty}f, D_{\infty}f)\) is called the attracting homological lamination of \(f\) (homological shadow of the attracting lamination). The proof essentially uses the fact that any fully irreducible automorphism \(f\) of a free group \(F_n\) has a train track representation \(\phi:G\to G\) (a continuous self-map on a graph \(G\) with nice properties with respect to \(\phi\)), and the transition matrix of \(\phi\) is Perron-Frobenius. The author first uses the half point graph \(HP_{\phi}\) of \(\phi\) and its variation to prove the existence of some version of attracting homological lamination and probability measure on \(\mathbb{R}^{E(G)}\), then he proves that the projection of the attracting homological lamination and probability measure on \(\mathbb{R}^{E(G)}\) provides the desired objects on \(H_1(F_n;\mathbb{R})\).
      0 references
      mapping class group
      0 references
      pseudo-Anosov map
      0 references
      \(\mathrm{Aut}(F_n)\)
      0 references
      0 references

      Identifiers