Table of Contents
Fetching ...

Motivic pieces of curves: $L$-functions and periods

Harry Spencer

TL;DR

The paper develops and tests a framework in which the $L$-function of a curve $C$ with a finite group action $G$ factors into motivic pieces $L(C^ au,s)$ attached to irreducible $G$-representations $ au$. A key contribution is Algorithm 3 (The algorithm), which reduces the computation of Euler factors for $L(C^ au,s)$ to counting points on auxiliary curves, enabling explicit numerical verifications in cases with $G=C_3$, $C_4$, and $D_{10}$, and allowing factorisations of Jacobians of GL$_2$-type via Hecke-type endomorphisms. The work further connects these motivic pieces to Deligne’s Period Conjecture, providing a practical method to compute Deligne periods for motivic pieces arising from superelliptic curves and giving substantial numerical evidence for the conjecture in exponent-three and exponent-four settings. Overall, this framework links explicit point-counting, modularity phenomena, and period conjectures to provide both computational tools and numerical validation for conjectures in arithmetic geometry related to motivic L-functions.

Abstract

Given a curve $C$ over a number field $K$ equipped with the action of a finite group $G$ by $K$-automorphisms, one obtains a factorisation of $L(C,s)$ into a product of $L$-functions of `motivic pieces of curves' associated to irreducible $G$-representations. We describe an algorithm for explicitly computing values of these $L$-functions, demonstrating implementations in the cases of certain curves with actions by $C_3$, $C_4$ and $D_{10}$. We explain how this algorithm can be used to factor $L$-functions of curves with endomorphisms of Hecke type. Towards applications, we explicitly formulate and numerically verify a version of Deligne's Period Conjecture for hitherto-uninvestigated $L$-functions arising from motivic pieces of superelliptic curves.

Motivic pieces of curves: $L$-functions and periods

TL;DR

The paper develops and tests a framework in which the -function of a curve with a finite group action factors into motivic pieces attached to irreducible -representations . A key contribution is Algorithm 3 (The algorithm), which reduces the computation of Euler factors for to counting points on auxiliary curves, enabling explicit numerical verifications in cases with , , and , and allowing factorisations of Jacobians of GL-type via Hecke-type endomorphisms. The work further connects these motivic pieces to Deligne’s Period Conjecture, providing a practical method to compute Deligne periods for motivic pieces arising from superelliptic curves and giving substantial numerical evidence for the conjecture in exponent-three and exponent-four settings. Overall, this framework links explicit point-counting, modularity phenomena, and period conjectures to provide both computational tools and numerical validation for conjectures in arithmetic geometry related to motivic L-functions.

Abstract

Given a curve over a number field equipped with the action of a finite group by -automorphisms, one obtains a factorisation of into a product of -functions of `motivic pieces of curves' associated to irreducible -representations. We describe an algorithm for explicitly computing values of these -functions, demonstrating implementations in the cases of certain curves with actions by , and . We explain how this algorithm can be used to factor -functions of curves with endomorphisms of Hecke type. Towards applications, we explicitly formulate and numerically verify a version of Deligne's Period Conjecture for hitherto-uninvestigated -functions arising from motivic pieces of superelliptic curves.
Paper Structure (14 sections, 13 theorems, 45 equations, 1 table, 1 algorithm)

This paper contains 14 sections, 13 theorems, 45 equations, 1 table, 1 algorithm.

Key Result

Theorem 1.4

For the $14$ genus $3$ superelliptic curves with $L(C,1)\ne0$ considered in §sec:examples, there exist $a,b,c\in\mathbb{Z}$ with $|a|,|b|\le 16$ and $|c|\le160$ such that where $\tau$ is a non-$\mathbb{Q}$-rational character of $C_3$ or $C_4$, $L^{\text{approx}}(C^\tau,1)$ is our numerical value for $L(C^\tau,1)$, $\Omega^\text{approx}_{C^\tau}$ is that for the Deligne period as in §sec:super_per

Theorems & Definitions (63)

  • Conjecture 1.2
  • Theorem 1.4
  • Remark 1.6
  • Remark 1.7
  • Example 1.8: = Example \ref{['ex:pic1']}
  • Remark 1.11
  • Proposition 1.12: = Proposition \ref{['prop:mod_form']}
  • Definition 2.1: = DGKM2
  • Definition 2.2
  • Remark 2.3
  • ...and 53 more