🔢 mathematics

On the Fields of Moduli of Curves of Genus Six

This paper investigates the conditions under which curves of genus six admit a model over their field of moduli by analyzing the stratification of the moduli space $M_6$ .

Original authors: Tianzhi Yang

Published 2026-07-22

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Tianzhi Yang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a vast, invisible library where every book is a unique geometric shape, a curve that twists and turns in a world of pure mathematics. Mathematicians call these shapes "curves," and they organize them into a giant catalog called a "moduli space." Think of this space like a massive map where every single dot represents a different curve. Now, here is the tricky part: these curves are often described using complex numbers, which are like a super-charged version of the numbers we use every day. But what if we wanted to describe one of these curves using only "real" numbers, the kind you can find on a ruler or a thermometer?

The central puzzle this paper tackles is about "fields of moduli." Imagine you have a secret recipe for a cake. You can describe the cake perfectly to a friend in a different language, and they can understand exactly what it looks like. That description is your "field of moduli." The big question is: Can you actually bake the cake using only ingredients available in your own kitchen (your "field of definition")? Sometimes, the answer is yes; sometimes, the recipe is so specific that you need a special ingredient from the other friend's kitchen to make it real. This paper investigates exactly when this "baking" is possible for a very specific, complex type of curve: those with a "genus" of six. In simple terms, genus is a count of the holes in a shape; a genus-6 curve is a complicated, six-holed donut.

The authors, led by Tianzhi Yang, dive deep into this six-holed world to see which of these curves can be "baked" in their home kitchen and which ones get stuck needing foreign ingredients. They don't just guess; they use a powerful new mathematical toolkit involving "gerbes," which are like invisible scaffolding that helps mathematicians see the hidden connections between a shape and its description. By breaking the world of genus-6 curves into different neighborhoods based on their shapes—some look like fancy plane drawings, others like twisted double-covers, and some sit on special surfaces called del Pezzo surfaces—they map out exactly where the "baking" works and where it fails.

The paper's main finding is a detailed map of success and failure. They prove that for several types of these curves, the answer is always "yes": you can always bake them at home. For instance, if the curve is "hyperelliptic" (a specific kind of twisted shape) and its symmetry group is complex enough, it can always be defined over its field of moduli. Similarly, if the curve is a "plane quintic" (a shape drawn on a flat surface with a specific complexity), it is guaranteed to be definable at home.

However, the paper also rules out the idea that every curve can be baked at home. They identify specific scenarios where the curve refuses to stay in its home kitchen. For curves that sit on a smooth, five-sided surface (a smooth del Pezzo surface), the authors show that the curve can only be defined over its field of moduli unless its symmetry group is extremely simple (specifically, a group of order 2) or a specific type of group called $D_{10}$ , and even then, only if a special number called $\sqrt{-1}$ (the square root of negative one) is missing from the kitchen. If that number is missing, the curve cannot be baked at home.

Furthermore, the paper explores curves sitting on "singular" surfaces (surfaces with kinks or sharp points). For most types of these kinks (labeled $A_4, A_3, A_2+A_1, A_2$ , and $A_1$ ), the authors prove with certainty that the curves can always be defined over their field of moduli. But there is one stubborn exception: the "2A1" type. In this specific case, the authors show that the mathematical machinery breaks down, and they cannot guarantee the curve can be baked at home. While they haven't found a concrete example of a curve that fails in this specific case yet, the math suggests it's possible, leaving a tiny door open for future discovery.

In the end, this paper doesn't just say "it depends." It draws a precise boundary line. It tells us that for the vast majority of six-holed curves, the recipe is safe to use at home. But for a few very specific, highly symmetric, or oddly kinked shapes, the universe demands a special ingredient that might not be in your local pantry. The authors have successfully categorized these exceptions, turning a vague question into a clear, proven map of where the magic of geometry can be recreated and where it remains just out of reach.

Technical Summary: On the Fields of Moduli of Curves of Genus Six

Problem Statement
The paper addresses the fundamental question in arithmetic geometry regarding the "field of moduli" versus the "field of definition" for algebraic varieties. Let $K$ be an algebraically closed field of characteristic 0 and $k$ a subfield. For a variety $X$ (potentially with additional structure $\xi$ ), the field of moduli $k(X, \xi)$ is defined as the fixed field of the subgroup of $\text{Gal}(K/k)$ that stabilizes the isomorphism class of $(X, \xi)$ . While every field of definition contains $k(X, \xi)$ , it is not guaranteed that $k(X, \xi)$ itself is a field of definition. The central problem is to determine under what conditions a curve $C$ of genus 6 admits a model over its field of moduli.

Methodology
The author employs the theory of algebraic gerbes, specifically the framework developed by Bresciani and Vistoli, to translate the descent problem into a question of neutrality.

Gerbe Formulation: The set of twisted forms of a curve $C$ is classified by a residue gerbe $\mathcal{G}_C$ over its field of moduli $k_C$ . The curve $C$ descends to $k_C$ if and only if $\mathcal{G}_C$ is neutral (i.e., possesses a $k_C$ -rational point).
Faithful Morphisms: To analyze the neutrality of $\mathcal{G}_C$ , the paper constructs faithful morphisms $\mathcal{G}_C \to B\Gamma$ , where $\Gamma$ is an algebraic group scheme defined over $k_C$ derived from "distinguished structures" intrinsic to the curve (structures preserved by all Galois-semilinear automorphisms).
Neutral Subgroups: The problem reduces to determining whether the image of the automorphism group $\text{Aut}(C)$ inside $\Gamma(K)$ constitutes a neutral subgroup. A finite subgroup $G \subset \Gamma(K)$ is neutral if every $G$ -gerbe over an extension of $k_C$ has a rational point.
Stratification: The moduli space $\mathcal{M}_6$ $M_{6}$ is stratified according to the geometric classification of genus 6 curves. The analysis is performed case-by-case on these strata:
- Hyperelliptic curves
- Trigonal curves
- Plane quintics
- Bielliptic curves
- Curves on smooth quintic del Pezzo surfaces
- Curves on singular quintic del Pezzo surfaces (ADE types)

Key Contributions and Results

Stratification and Distinguished Structures: The paper systematically identifies the distinguished algebraic structures for each stratum of $\mathcal{M}_6$ and constructs the corresponding faithful morphisms to classifying stacks $B\Gamma$ :
- Hyperelliptic/Trigonal: Morphisms to $B\text{PGL}_2$ .
- Plane Quintics: Morphisms to $B\text{PGL}_3$ .
- Bielliptic: The unique bielliptic involution splits the Hodge bundle, yielding a morphism to $B(\mathbb{G}_m \times \text{GL}_5)$ .
- Del Pezzo Surfaces: The unique del Pezzo surface $S_C$ containing the curve (smooth or singular) provides a morphism to $B\text{Aut}(S_C)$ . For smooth surfaces, $\Gamma \cong S_5$ ; for singular surfaces, $\Gamma$ is the automorphism group of a split model of the specific ADE singularity type.
Classification of Neutral Subgroups in $S_5$ : A significant technical component is the classification of neutral subgroups of the symmetric group $S_5$ . The paper proves that, up to conjugacy, every subgroup of $S_5$ is neutral except:
1. The subgroup generated by a double transposition ( $C_2 \cong \langle (12)(34) \rangle$ ), which is never neutral.
2. The transitive subgroup $D_{10}$ , which is neutral if and only if $\sqrt{-1} \in k$ .
Main Theorems:
- Theorem A (Known/Partial Cases):
  - Hyperelliptic curves descend if $\text{Aut}(C)/\langle \iota \rangle$ is non-cyclic.
  - Plane quintics always descend.
  - For trigonal curves, if they do not descend, $\text{Aut}(C)$ is cyclic or a $C_3$ -extension of a cyclic group. The paper provides a simplified proof for this case using the uniqueness of the trigonal pencil.
- Theorem B (New Results):
  - Bielliptic curves: Always descend to their field of moduli. The proof utilizes the splitting of the Hodge bundle and shows the relevant representation is neutral.
  - Smooth Del Pezzo case: A curve $C$ on a smooth quintic del Pezzo surface descends unless $\text{Aut}(C) \cong C_2$ or ( $\text{Aut}(C) \cong D_{10}$ and $\sqrt{-1} \notin k$ ).
  - Singular Del Pezzo case: If the surface $S_C$ has a singularity type other than $2A_1$ (specifically $A_4, A_3, A_2+A_1, A_2, A_1$ ), the curve descends. This relies on the fact that the automorphism groups for these types are split connected solvable groups, for which all finite subgroups are neutral.
  - Exceptional Case ( $2A_1$ ): The paper identifies the singularity type $2A_1$ as the sole remaining exceptional case where the automorphism group is $G_m^2 \rtimes C_2$ . The paper notes that for this case, the normalizer quotient may fail to be special, and counterexamples (non-descending curves) are likely to exist, though none are explicitly constructed in the text.

Significance and Claims
The paper claims to provide a comprehensive analysis of the field of moduli problem for genus 6 curves by leveraging the gerbe-theoretic approach. Its primary significance lies in:

Completing the Classification: It resolves the descent problem for the bielliptic case and the majority of the del Pezzo cases, which were previously open or only partially understood.
Unifying Framework: It demonstrates the efficacy of the neutral subgroup framework (developed in [BY26]) in handling diverse geometric situations, from hyperelliptic to non-special curves.
Refining Counterexamples: It clarifies the conditions under which counterexamples exist (specifically linking the non-descent of $D_{10}$ -symmetric curves to the absence of $\sqrt{-1}$ in the base field) and isolates the $2A_1$ singularity type as the primary locus for potential new counterexamples.

The author explicitly states that no elementary proofs are known for some related results (e.g., real coefficients for odd-degree plane curves) and that the gerbe approach provides the necessary machinery to handle these descent questions. The paper does not claim to solve the problem for the $2A_1$ case but identifies it as the boundary of current solvability within this framework.

More like this