Table of Contents
Fetching ...

A uniform bound on the smallest surjective prime of an elliptic curve

Tyler Genao, Jacob Mayle, Jeremy Rouse

TL;DR

The paper proves that for any non-CM elliptic curve $E/\mathbb{Q}$, the smallest prime $\ell$ with surjective $\ell$-adic Galois representation satisfies $\ell\le 7$, and completely classifies curves with $\ell=7$ as the smallest surjective prime. The authors develop a framework using modular curves $X_H$ and fiber products to reduce the problem to finite rational points on a small set of curves, then compute these points with Magma to identify a finite set of exceptional $j$-invariants. They show that, except for six specific $j$-invariants, the smallest surjective prime is at most $5$, and they exhibit explicit curves (including an example with smallest surjective prime $7$) where the bound is sharp. Consequently, at least one of the $2$-, $3$-, or $5$-adic representations is surjective for every non-CM $E/\mathbb{Q}$, providing a precise, computationally verifiable instance of Serre’s open image theme and contributing toward a uniform bound on surjective primes in the non-CM case. The work combines modular-curve techniques with explicit arithmetic geometry computations to yield a concrete, finite classification and an accessible computational roadmap.

Abstract

Let $E/\mathbb{Q}$ be an elliptic curve without complex multiplication. A well-known theorem of Serre asserts that the $\ell$-adic Galois representation $ρ_{E,\ell^\infty}$ is surjective for all but finitely many prime numbers $\ell$. Considerable work has gone into bounding the largest possible nonsurjective prime; a uniform bound of $37$ has been proposed but is yet unproven. We consider an opposing direction, proving that the smallest prime $\ell$ such that $ρ_{E,\ell^\infty}$ is surjective is at most $7$. Moreover, we completely classify all elliptic curves $E/\mathbb{Q}$ for which the smallest surjective prime is exactly $7$.

A uniform bound on the smallest surjective prime of an elliptic curve

TL;DR

The paper proves that for any non-CM elliptic curve , the smallest prime with surjective -adic Galois representation satisfies , and completely classifies curves with as the smallest surjective prime. The authors develop a framework using modular curves and fiber products to reduce the problem to finite rational points on a small set of curves, then compute these points with Magma to identify a finite set of exceptional -invariants. They show that, except for six specific -invariants, the smallest surjective prime is at most , and they exhibit explicit curves (including an example with smallest surjective prime ) where the bound is sharp. Consequently, at least one of the -, -, or -adic representations is surjective for every non-CM , providing a precise, computationally verifiable instance of Serre’s open image theme and contributing toward a uniform bound on surjective primes in the non-CM case. The work combines modular-curve techniques with explicit arithmetic geometry computations to yield a concrete, finite classification and an accessible computational roadmap.

Abstract

Let be an elliptic curve without complex multiplication. A well-known theorem of Serre asserts that the -adic Galois representation is surjective for all but finitely many prime numbers . Considerable work has gone into bounding the largest possible nonsurjective prime; a uniform bound of has been proposed but is yet unproven. We consider an opposing direction, proving that the smallest prime such that is surjective is at most . Moreover, we completely classify all elliptic curves for which the smallest surjective prime is exactly .
Paper Structure (4 sections, 3 theorems, 13 equations, 1 table)

This paper contains 4 sections, 3 theorems, 13 equations, 1 table.

Key Result

Theorem 1

If $E/\mathbb{Q}$ is a non-CM elliptic curve, then $\rho_{E,\ell^\infty}$ is surjective for all sufficiently large prime numbers $\ell$.

Theorems & Definitions (6)

  • Theorem 1: Serre MR387283, 1972
  • Theorem 2
  • Example 3
  • Example 4
  • Lemma 5
  • proof