Table of Contents
Fetching ...

Lower bounds for heights on some algebraic dynamical systems

Arnaud Plessis, Satyabrat Sahoo

TL;DR

The article develops lower bounds for heights in algebraic dynamical systems, connecting three main settings: the Weil height on P^1, the Néron–Tate height on abelian varieties, and dynamical heights. It generalizes Pottmeyer’s elliptic-curve bound to simple abelian varieties degenerate at a place, yielding a $\\hat{h}$-lower bound in terms of $l_v(P)$ and highlighting Lehmer-type consequences; it then proves a tropical version of a height bound via Yamaki’s lemma and Poincaré reducibility, providing a broad criterion requiring nontrivial tropicalizations of simple factors. Building on this, the paper gives a dynamical Bogomolov theorem: if the Julia set contains an atypical element, then $\\mathbb{P}^1(K^{nr,v})$ has the strong Bogomolov property for $\\hat{h}_φ$, with explicit Newton-polygon criteria enabling concrete polynomial families to satisfy the hypothesis. The results combine tropical geometry, Berkovich dynamics, and equidistribution to yield new Lehmer-type bounds and dynamical Bogomolov statements with potential applications to arithmetic dynamics and Diophantine geometry.

Abstract

Let $v$ be a finite place of a number field $K$ and write $K^{nr,v}$ for the maximal field extension of $K$ in which $v$ is unramified. The purpose of this paper is split up into two parts. The first one generalizes a theorem of Pottmeyer: If $E$ is an elliptic curve defined over $K$ with split multiplicative reduction at $v$, then the Néron-Tate height of a non-torsion point $P\in E(\bar{K})$ is bounded from below by $C / e_v(P)^{2 e_v(P)+1}$, where $C>0$ is an absolute constant and $e_v(P)$ is the maximum of all ramification indices $e_w(K(P) \vert K)$ with $w\vert v$. Among other things, we refine this result by showing that given a simple abelian variety $A$ defined over $K$ that is degenerate at $v$, the Néron-Tate height of a non-torsion point $P\in A(\bar{K})$ is at least $C / \mathrm{lcm}_{w\vert v} \{e_w(K(P)\vert K)\}^2$, where $C>0$ is an absolute constant. We then give applications towards Lehmer's conjecture. Next, we provide the first examples of polynomials $φ\in K[X]$ of degree at least $2$ so that the canonical height $\hat{h}_φ$ of any point in $\bbP^1(K^{nr,v})$ is either $0$ or bounded from below by an absolute positive constant.

Lower bounds for heights on some algebraic dynamical systems

TL;DR

The article develops lower bounds for heights in algebraic dynamical systems, connecting three main settings: the Weil height on P^1, the Néron–Tate height on abelian varieties, and dynamical heights. It generalizes Pottmeyer’s elliptic-curve bound to simple abelian varieties degenerate at a place, yielding a -lower bound in terms of and highlighting Lehmer-type consequences; it then proves a tropical version of a height bound via Yamaki’s lemma and Poincaré reducibility, providing a broad criterion requiring nontrivial tropicalizations of simple factors. Building on this, the paper gives a dynamical Bogomolov theorem: if the Julia set contains an atypical element, then has the strong Bogomolov property for , with explicit Newton-polygon criteria enabling concrete polynomial families to satisfy the hypothesis. The results combine tropical geometry, Berkovich dynamics, and equidistribution to yield new Lehmer-type bounds and dynamical Bogomolov statements with potential applications to arithmetic dynamics and Diophantine geometry.

Abstract

Let be a finite place of a number field and write for the maximal field extension of in which is unramified. The purpose of this paper is split up into two parts. The first one generalizes a theorem of Pottmeyer: If is an elliptic curve defined over with split multiplicative reduction at , then the Néron-Tate height of a non-torsion point is bounded from below by , where is an absolute constant and is the maximum of all ramification indices with . Among other things, we refine this result by showing that given a simple abelian variety defined over that is degenerate at , the Néron-Tate height of a non-torsion point is at least , where is an absolute constant. We then give applications towards Lehmer's conjecture. Next, we provide the first examples of polynomials of degree at least so that the canonical height of any point in is either or bounded from below by an absolute positive constant.

Paper Structure

This paper contains 16 sections, 14 theorems, 30 equations.

Key Result

Theorem 1.1

Let $E$ be an elliptic curve defined over a number field $K$ with split multiplicative reduction at some finite place $v$ of $K$. Let $\mathcal{L}$ be a symmetric ample line bundle on $E$. Then there is an effective computable constant $C>0$ such that for all non-torsion points $P\in E(\bar{K})$, where $e_v(P)$ denotes the maximum of all ramification indices $e_w(K(P)\vert K)$ with $w$ a place of

Theorems & Definitions (32)

  • Theorem 1.1: Pottmeyer, P15, Theorem 4.1
  • Theorem 1.2
  • Remark 1.3
  • Remark 1.4
  • Remark 1.5
  • Remark 1.6
  • Conjecture 1.7: Lehmer's conjecture
  • Corollary 1.8
  • Remark 1.9
  • Remark 1.10
  • ...and 22 more