On the uniform equidistribution of long closed horocycles (Q6230821): Difference between revisions

From MaRDI portal
Added link to MaRDI item.
Set OpenAlex properties.
 
Property / full work available at URL
 
Property / full work available at URL: https://doi.org/10.1215/s0012-7094-04-12334-6 / rank
 
Normal rank
Property / OpenAlex ID
 
Property / OpenAlex ID: W2011423676 / rank
 
Normal rank

Latest revision as of 09:46, 30 July 2024

scientific article; zbMATH DE number 2103765
Language Label Description Also known as
English
On the uniform equidistribution of long closed horocycles
scientific article; zbMATH DE number 2103765

    Statements

    On the uniform equidistribution of long closed horocycles (English)
    0 references
    27 September 2004
    0 references
    Given a hyperbolic surface of finite area with cusps, a horocycle is a level set of the Busemann function associated to a ray go in into the cusp. When we are inside the cusp, the horocycle is a simple closed curve that becomes arbitrarily short when we go to the corresponding end of the manifold. If we move in the other direction, the horocycles become arbitrarily long and are not simple curves anymore. In fact, they become equidistributed when the length goes to infinity. Namely, if \(l\) denotes the length or the horocycle, and \(A\) any regular subset of the hyperbolic surface, the ratio between the length of the intersection of the horocycle with \(A\) and the total length \(l\) converges to the ratio between the area of \(A\) and the area of the surface. The main theorem of this paper shows that the equidistribution holds for subsegments of the horocycle provided that the length is larger than \(l^{1/2+\varepsilon}\). The power \(1/2+\varepsilon\) is the best possible: segments of length \(l^{1/2}\) can be not equidistributed. This improves previous versions of this result: for instance, \textit{D. A. Hejhal} [Asian J. Math. 4, 839--854 (2000; Zbl 1014.11038)] proved this for some power depending on the group, and a proof when the length is precisely equal to \(l^{1/2+\varepsilon}\) for some fixed exponent \(1/2+\varepsilon\) can be deduced from the techniques of \textit{N. A. Shah} [Proc. Indian Acad. Sci., Math. Sci. 106, 105--125 (1996; Zbl 0864.22004)] and \textit{A. Eskin} and \textit{C. McMullen} [Duke Math. J. 71, 181--209 (1993; Zbl 0798.11025)]. If one replaces sets by tests functions, this theorem is read in terms of functions and two proofs are given, one using ergodic theory and the other one spectral theory. The former allows one to obtain asymptotic equidistribution on the unit tangent bundle, the latter one gives explicit information of the rate of convergence. There is also a result on the asymptotic joint equidistribution of a finite number of distinct subsegments having length proportional to the length of the horocycle.
    0 references
    0 references
    Fuchsian group
    0 references
    horocycle
    0 references
    horocyclic flow
    0 references
    asymptotic equidistribution
    0 references
    ergodicity
    0 references
    spectral theory
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references