Carathéodory convergence and the conformal type problem
Alexandre Eremenko, Sergei Merenkov · 2024
Read Carathéodory convergence and the conformal type problem by Alexandre Eremenko, Sergei Merenkov (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-05995.txt.
Introduction
Alexandre Eremenko and Sergei Merenkov study a convergence idea for curved surfaces in a 2024 paper. The surfaces are open, simply connected, and spread over a sphere. Their focus is the conformal type problem, which asks what basic shape class such a surface falls into. The pair give cases where that class can change inside the Speiser class when two special values merge. The short paper adds a caution to a classic corner of complex analysis.
Catalog record
| CREATOR | Alexandre Eremenko, Sergei Merenkov |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-05995
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Carathéodory convergence and the conformal type problem Thanks: S. M. supported by NSF grant DMS-2247364. Alexandre Eremenko, Sergei Merenkov Abstract . We study Carathéodory convergence for open, simply connected surfaces spread over the sphere and, in particular, provide examples demonstrating that in the Speiser class the conformal type can change when two singular values collide. 1. Introduction Carathéodory Kernel Convergence is an important tool in the theory of univalent functions. It gives a geometric criterion for a sequence of normalized univalent functions in the unit disk to converge uniformly on compacta, and gives a description of the image of the disk under the limiting map. In this paper we adapt the notion of convergence in the sense of Carathéodory, introduced in C. Carathéodory [ Ca12 ] , see also L. I. Volkovyskiĭ [ Vo48 ] , to the setting of pointed surfaces spread over the sphere. Moreover, we establish results, Theorems 3.1 , 3.3 , that relate such convergence to convergence on compacta omitting certain exceptional sets. A result similar to the necessary part of Theorem 3.1 for surfaces spread over the plane was proved by K. Biswas and R. Perez-Marco [ BPM15 , Theorem 1.2] . Another aim of this paper is to provide examples of sequences of open, simply connected surfaces spread over the sphere (in fact, over the plane) that have only finitely many singular values and whose conformal type changes when two of the singular values collide; see Section 4 . We also give an example of a sequence of entire functions in the plane with finitely many values, so that each function in the sequence has infinite order, while the limit has order one; see Section 5 . 1.1. Surfaces spread over the sphere Classically, Riemann surfaces are thought of as surfaces associated to holomorphic or, more generally, meromorphic functions. Definition 1.1 . A surface spread over the sphere is a pair (S,f) , where S is an open topological surface and f\colon S\to\overline{\mathbb{C}} is a continuous, open and discrete map, called a projection. Here, \overline{\mathbb{C}} is the Riemann sphere. The surface S can be endowed with the pull-back conformal structure, so that f becomes holomorphic. In what follows, we do not distinguish two surfaces (S_{1},f_{1}) and (S_{2},f_{2}) if there exists a homeomorphism h\colon S_{1}\to S_{2} such that f_{1}=f_{2}\circ h. The homeomorphism h is conformal when S_{1} and S_{2} are endowed with the pull-back conformal structures. If S is an open, simply connected surface, then, equipped with the pull-back conformal structure, it is equivalent to either the complex plane {\mathbb{C}} or the unit disk {\mathbb{D}} in {\mathbb{C}} . In the former case we call (S,f) parabolic , and in the latter hyperbolic . For a survey on surfaces spread over the sphere and the type problem one can consult [ Er21 ] . Near each point, a continuous, open and discrete map f is homeomorphically equivalent to the map z\mapsto z^{d} , where d\in{\mathbb{N}} . More precisely, for each p_{0}\in S , there exists an open neighborhood U of p_{0} and two homeomorphisms h_{1},h_{2} , such that h_{1}\colon U\to{\mathbb{D}},\ h_{2}\colon f(U)\to{\mathbb{D}} , with h_{1}(p_{0})=0,\ h_{2}(f(p_{0}))=0 , and h_{2}\circ f\circ h_{1}^{-1}(z)=z^{d},\quad z\in{\mathbb{D}}. The number d is called the local degree of f at p_{0} . It does not depend on the choice of homeomorphisms h_{1} and h_{2} . If d>1 , p_{0} is called the critical point and f(p_{0}) the critical value of f . An element a\in\overline{\mathbb{C}} is called an asymptotic value of f if there exists a path \gamma\colon[0,1)\to S , called an asymptotic tract , that leaves every compact set of S as t\to 1 , and such that \lim_{t\to 1}f(\gamma(t))=a. A singular value of f is either a critical or an asymptotic value. Definition 1.2 . A surface spread over the sphere (S,f) is said to belong to Speiser class \mathcal{S} , if the projection map f has only finitely many singular values. Examples of surfaces from Speiser class include ({\mathbb{C}},p) , where p is an arbitrary polynomial, ({\mathbb{C}},\exp) , ({\mathbb{C}},\sin) , ({\mathbb{C}},\cos) , ({\mathbb{C}},\exp\circ\exp) , ({\mathbb{C}},\wp) , where \wp is the Weierstrass \wp -function, ({\mathbb{D}},\lambda) , where \lambda is the modular function, etc. Surfaces from Speiser class have combinatorial descriptions in terms of labeled Speiser graphs as follows. Let \beta be a base curve , i.e., a curve in the sphere that contains all singular values of f . For example, when all singular values of f are real or \infty , we can chose \beta to be the extended real line. Then, \beta divides the sphere \overline{\mathbb{C}} into two topological hemispheres, which we can call an “upper” and “lower” hemispheres for convenience. We fix two points, one in each of the hemispheres, and denote the point in the upper hemisphere by \times and the one in the lower hemisphere by \circ . If the number of singular values of f is k , we look at the graph G_{\beta} in \overline{\mathbb{C}} with two vertices \times and \circ and k edges, each edge connecting \times to \circ in such a way that it separates two adjacent singular values on \beta . The Speiser graph G_{f,\beta} of (S,f) is then the preimage of G_{\beta} under the map f . It is bipartite, homogeneous of degree k , and its faces, i.e., connected components of the complement in S , are labeled by the corresponding singular values of f . The labels appear in cyclic orders around each vertex of G_{f,\beta} , according to the order of the singular values on \beta viewed from \times or \circ , respectively. See Figures 1 , 2 , 3 for examples. Conversely, given such a labeled graph G in the plane and a base curve \beta that contains all the labels of the faces, one can reconstruct a surface (S,f) by gluing together upper and lower hemispheres of \overline{\mathbb{C}}\setminus\{\beta\} identifying them along the arcs of \beta between adjacent labels according to the combinatorics of the graph G . See [ GO70 ] or [ Ne70 ] for further details. 1.2. Carathéodory convergence We consider triples T=(S,f,w) , where (S,f) is a surface spread over the sphere, and w\in S a point which is not critical for f . We refer to these triples as pointed surfaces spread over the sphere. Two pointed surfaces spread over the sphere T_{1}=(S_{1},f_{1},w_{1}) and T_{2}=(S_{2},f_{2},w_{2}) are equivalent , denoted T_{1}\sim T_{2} , if there exists a homeomorphism h\colon S_{1}\to S_{2} such that h(w_{1})=w_{2} and f_{1}=f_{2}\circ h . Equivalence classes are still called pointed surfaces spread over the sphere . In addition to the equivalence, we define the order relation on surfaces spread over the sphere: (S_{1},f_{1},w_{1})\subset(S_{2},f_{2},w_{2}) means that there is a continuous injective map \phi\colon S_{1}\to S_{2} such that (1) \phi(w_{1})=w_{2}\quad\mbox{and}\quad f_{1}=f_{2}\circ\phi. The second equation in ( 1 ) implies that \phi is holomorphic. It is easy to see that T_{1}\subset T_{2} and T_{2}\subset T_{1} imply that there is a homeomorphism h\colon S_{1}\to S_{2} satisfying ( 1 ) with \phi=h . In this case T_{1}\sim T_{2} . Each equivalence class contains a normalized triple with S=D_{R}:=\{z\in{\mathbb{C}}:|z|<R\} for some R\in(0,+\infty] , w=0 , and f^{\#}(0)=1 , where f^{\#} is the spherical derivative, f^{\#}=\frac{f^{\prime}}{1+|f|^{2}}. A triple T is called maximal if T\subset T_{1}