Table of Contents
Fetching ...

Length minimization of filling pairs on hyperbolic surfaces

Ni An, Bhola Nath Saha, Bidyut Sanki

TL;DR

This work classifies minimal filling pairs on genus-two surfaces up to mapping class group actions, showing there are exactly two types, $\{4,12\}$ and $\{8,8\}$. It then derives sharp lower bounds on the length of such pairs: $\ge 8\,\mathrm{cosh}^{-1}(\sqrt{2}+1)$ for the $\{8,8\}$ type and $\ge L_0$ with $L_0=6\,\mathrm{cosh}^{-1}(7/2)$ for the $\{4,12\}$ type, with the global minimum realized in the latter. The $\{4,12\}$ bound is obtained via a detailed quadrilateral construction and constrained optimization (Lagrange multipliers), supported by numerical verification and symmetry arguments; the $\{8,8\}$ bound leverages Gauss–Bonnet and isoperimetric inequalities for hyperbolic polygons. Collectively, these results establish a concrete, global lower bound on the length of any filling pair in genus two, contributing to extremal problems in Teichmüller theory and hyperbolic geometry.

Abstract

A filling pair $(α, β)$ of a surface $S_g$ is a pair of simple closed curves in minimal position such that the complement of $α\cupβ$ in $S_g$ is a disjoint union of topological disks. A filling pair is said to be minimally intersecting if the number of intersections between them, or equivalently, the number of complementary disks, is minimal among all filling pairs of $S_g$. For surfaces of genus $g \geq 3$, minimal filling pairs are well understood, whereas in genus two, such a pair divides the surface into exactly two disks. In this paper, we classify all minimal filling pairs up to the action of the mapping class group in genus two and determine the length of the shortest minimal filling pair.

Length minimization of filling pairs on hyperbolic surfaces

TL;DR

This work classifies minimal filling pairs on genus-two surfaces up to mapping class group actions, showing there are exactly two types, and . It then derives sharp lower bounds on the length of such pairs: for the type and with for the type, with the global minimum realized in the latter. The bound is obtained via a detailed quadrilateral construction and constrained optimization (Lagrange multipliers), supported by numerical verification and symmetry arguments; the bound leverages Gauss–Bonnet and isoperimetric inequalities for hyperbolic polygons. Collectively, these results establish a concrete, global lower bound on the length of any filling pair in genus two, contributing to extremal problems in Teichmüller theory and hyperbolic geometry.

Abstract

A filling pair of a surface is a pair of simple closed curves in minimal position such that the complement of in is a disjoint union of topological disks. A filling pair is said to be minimally intersecting if the number of intersections between them, or equivalently, the number of complementary disks, is minimal among all filling pairs of . For surfaces of genus , minimal filling pairs are well understood, whereas in genus two, such a pair divides the surface into exactly two disks. In this paper, we classify all minimal filling pairs up to the action of the mapping class group in genus two and determine the length of the shortest minimal filling pair.
Paper Structure (6 sections, 14 theorems, 25 equations, 9 figures)

This paper contains 6 sections, 14 theorems, 25 equations, 9 figures.

Key Result

Theorem 1.1

A minimally intersecting filling pair $(\alpha, \beta)$ of a closed surface $S_2$ of genus two is of either type $\{4, 12\}$ or type $\{8,8\}$ (equivalently, there does not exist a filling pair of type $\{6,10\}$ on $S_2$).

Figures (9)

  • Figure 2.1: Local picture of the surface obtained from a fat graph.
  • Figure 2.2: The boundary of the surface is oriented so that the surface part lies on the right side.
  • Figure 2.3: The action of $\sigma_0\sigma_1$.
  • Figure 4.1:
  • Figure 4.2:
  • ...and 4 more figures

Theorems & Definitions (26)

  • Theorem 1.1
  • Theorem 1.2
  • Theorem 1.3
  • Theorem 1.4
  • Definition 2.1: Graph
  • Definition 2.2
  • Definition 2.3: Standard cycle
  • Lemma 2.4
  • proof
  • Theorem 3.1
  • ...and 16 more