The natural flow and the critical exponent

Chris Connell,D. B. McReynolds,Shi Wang

2024-09-04

Abstract:Inspired by work of Besson-Courtois-Gallot, we construct a flow called the natural flow on a nonpositively curved Riemannian manifold $M$. As with the natural map, the $k$-Jacobian of the natural flow is directly related to the critical exponent $\delta$ of the fundamental group. There are several applications of the natural flow that connect dynamical, geometrical, and topological invariants of the manifold. First, we give $k$-dimensional linear isoperimetric inequalities when $k > \delta$. This, in turn, produces lower bounds on the Cheeger constant. We resolve a recent conjecture of Dey-Kapovich on the non-existence of $k$-dimensional compact, complex subvarieties of complex hyperbolic manifolds with $2k > \delta$. We also provide upper bounds on the homological dimension, generalizing work of Kapovich and work of Farb with the first two authors. Using different methods, we also give upper bounds on the cohomological dimension which partially resolves a conjecture of Kapovich. We also specialize to the setting of locally symmetric manifolds where we further refine our work. Finally, we introduce a new growth condition on the Bowen-Margulis measure that we call uniformly exponentially bounded that we connect to the cohomological dimension and which could be of independent interest.

Differential Geometry,Dynamical Systems,Group Theory,Geometric Topology

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to explore the connections between different invariants on non - positive curvature Riemannian manifolds \(M\), and these invariants include dynamical, geometrical and topological invariants. Specifically, the author studies these problems by constructing a dynamical system called the natural flow. The natural flow is closely related to the critical exponent \(\delta\) of the fundamental group of the manifold, and the critical exponent can be used to control various different geometric and topological properties of the manifold \(M\). The following are some specific problems explored in the paper:

1. **Linear isoperimetric inequality**:
- If \(k>\delta\), then \(M\) satisfies the \(k\)-dimensional linear isoperimetric inequality. This provides a lower bound for the Cheeger constant in the case of codimension one.
- See Theorem 1.3 and Corollary 1.4 for specific results.

2. **Existence of complex hypersurfaces**:
- Solves a conjecture of Dey - Kapovich, that is, if \(M\) is a complete complex hyperbolic \(n\)-manifold and \(\delta(M)<2k\), then there are no compact \(k\)-dimensional complex sub - varieties in \(M\).
- See Theorem 1.5 for specific results.

3. **Homological dimension**:
- Provides upper bounds for the homological dimension, and these upper bounds depend on the critical exponent \(\delta\). This generalizes the work of Kapovich, Farb et al.
- See Theorem 1.6 and Corollaries 1.7, 1.8 for specific results.

4. **Cohomological dimension**:
- Using Morse theory and Patterson - Sullivan theory, provides upper bounds for the cohomological dimension, partially solving a conjecture of Kapovich.
- See Theorem 1.14 and Corollary 1.16 for specific results.

5. **Locally symmetric spaces**:
- Further refines the above work in the setting of locally symmetric spaces, especially for locally symmetric spaces of rank one and higher rank.
- See Section 4 for specific results.

6. **Growth conditions of Bowen - Margulis measure**:
- Introduces a new growth condition, called uniform exponential boundedness, and relates it to the cohomological dimension, and this condition may be of independent interest.
- See Theorem 1.15 for specific results.

The core tool in the paper is the natural flow. Through this flow, the author is able to relate the critical exponent \(\delta\) to various geometric and topological properties of the manifold, thus obtaining a series of important conclusions.