🔢 mathematics

Pryms of $\mathbb{Z}_3\times\mathbb{Z}_3$ coverings of genus 2 curves

This paper establishes that unramified Galois $\mathbb{Z}_3 \times \mathbb{Z}_3$ coverings of genus 2 curves are uniquely determined by their Prym varieties and investigates the generic finiteness of Prym maps for unramified coverings of such curves by arbitrary abelian groups.

Original authors: Paweł Borówka, Anatoli Shatsila

Published 2026-02-24

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Paweł Borówka, Anatoli Shatsila

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 you are a detective trying to solve a mystery, but instead of fingerprints or DNA, your clues are shapes and symmetries.

This paper is about a specific type of mathematical detective work involving curves (think of them as twisted loops or rubber bands) and coverings (like wrapping a complex, multi-layered blanket over a simple loop).

Here is the story of what the authors, Paweł Borówka and Anatoli Shatsila, discovered, explained in everyday language.

The Setup: The Mystery of the "Prym"

The Base Curve (The Mystery): Imagine a simple, twisted loop called a "genus 2 curve." It's a bit like a figure-eight shape.
The Covering (The Blanket): Now, imagine wrapping a much more complex, multi-layered sheet over this figure-eight. This sheet is a "covering." In this paper, the authors are looking at a very specific, highly symmetrical way of wrapping this sheet. They call it a $\mathbb{Z}_3 \times \mathbb{Z}_3$ covering.
- Analogy: Think of the base curve as a single wire. The covering is a complex, 9-stranded rope braided around it. The "braiding pattern" is determined by a group of symmetries (like rotating the rope in specific ways).
The Clue (The Prym Variety): When you wrap this complex rope, it leaves a "shadow" or a "fingerprint" on a special mathematical object called a Prym variety.
- Analogy: If you shine a light through a complex stained-glass window, the shadow it casts on the wall is the Prym variety. The shadow is a simplified version of the window, but it holds the key to what the window looks like.

The Big Question: Can You Reconstruct the Window from the Shadow?

The main question the authors ask is: If I only show you the shadow (the Prym variety), can you figure out exactly what the original complex rope (the covering) looked like?

In mathematics, this is called the Prym-Torelli Theorem.

The Answer: YES!
The Discovery: The authors proved that for this specific type of 9-stranded braided rope, the shadow is unique. If two different ropes cast the exact same shadow, they must be the same rope. You can perfectly reconstruct the original mystery just by looking at the clue.

How Did They Solve It? (The Detective Work)

The proof is like a multi-step puzzle. Here is how they did it, using simple analogies:

Step 1: Finding the Hidden Patterns
The shadow (Prym variety) has a specific texture (called a "polarization"). The authors showed that this texture contains hidden instructions. By reading these instructions, they could identify the "symmetry group" (the rules of how the rope was braided).

Analogy: It's like looking at a fingerprint and realizing, "Ah, this pattern only exists if the person was wearing a specific type of glove."

Step 2: Breaking it Down into Smaller Pieces
The complex rope isn't just one big mess; it's made of smaller, simpler loops. The authors showed that the shadow is actually a combination of four smaller shadows, each corresponding to a simpler part of the rope.

Analogy: Imagine the big shadow is a mosaic made of four smaller tiles. They figured out how to separate the tiles.

Step 3: Finding the "Elliptic" Keys
Inside each of those four smaller tiles, there are even simpler shapes called elliptic curves (which look like donuts). The authors proved that the texture of the shadow uniquely identifies exactly three of these donuts in each tile.

Analogy: They found that every tile has three specific "keyholes" that are unique to that tile.

Step 4: Reassembling the Puzzle
By matching up these keyholes (the elliptic curves) across the four tiles, they could figure out how the pieces fit together. They found that these pieces form a larger, intermediate shape (a genus 4 curve).

Analogy: Once they matched the keyholes, the four tiles snapped together to reveal a picture of a larger, simpler loop.

Step 5: The Final Reveal
From this intermediate loop, they could mathematically "unwind" the braiding to find the original complex rope and the base figure-eight.

Conclusion: The shadow (Prym variety) contains enough information to rebuild the entire original structure, piece by piece.

The Second Part: When Does This Work for Other Ropes?

The authors didn't just stop at this one type of rope. They asked: "Does this work for other braiding patterns?"

They looked at all possible ways to wrap a rope around a figure-eight using different symmetry groups (different numbers of strands and different braiding rules).

The Rule of Thumb:
- If the rope is very simple (like a 2-strand, 3-strand, 4-strand, or 5-strand braid), the shadow is not unique. Many different ropes can cast the same shadow. It's like having a blurry photo where you can't tell which car it is.
- If the rope is complex enough (6 strands or more, or a complex 2x2 or 3x3 pattern), the shadow becomes unique. The photo is sharp, and you can identify the car perfectly.

The Exception:
They found that if the "complexity" of the rope is too low (specifically, if the group of symmetries is a simple cycle with fewer than 6 steps), you lose information. But as soon as you get to more complex symmetries (like the 9-strand one they studied, or groups with 5 strands), the mystery is solvable.

Summary

The Problem: Can you identify a complex, braided rope just by looking at its shadow?
The Result: Yes, for a specific, complex 9-strand braid, the answer is a definite YES. The shadow is a perfect blueprint.
The Method: They broke the shadow down into smaller, recognizable pieces (elliptic curves) and used those pieces to reconstruct the whole.
The Bigger Picture: They also found that this "perfect reconstruction" works for almost all complex braids, but fails for very simple ones.

In short, the authors built a mathematical machine that can take a "shadow" of a complex geometric shape and perfectly reverse-engineer the original object, proving that in this specific world of curves, nothing is ever truly lost in the shadow.

1. Problem Statement

The paper investigates the geometry of Prym maps associated with unramified Galois coverings of smooth genus 2 curves ( $H$ ) by curves ( $\tilde{C}$ ) with an abelian Galois group. Specifically, the authors focus on two main problems:

Global Injectivity for $\mathbb{Z}_3 \times \mathbb{Z}_3$ : Determining whether the Prym map is injective for unramified coverings with Galois group $G \cong \mathbb{Z}_3 \times \mathbb{Z}_3$ . This involves reconstructing the covering curve and the map from the polarized Prym variety.
Generic Finiteness for General Abelian Groups: Characterizing the behavior of the Prym map for arbitrary finite abelian groups $G$ . The authors seek to identify for which groups the generic fiber of the Prym map is finite (i.e., the map is generically finite) versus when it has positive-dimensional fibers.

2. Methodology

The authors employ a combination of algebraic geometry, the theory of abelian varieties, and group actions on curves.

A. Moduli Space Decomposition

The moduli space of $\mathbb{Z}_3 \times \mathbb{Z}_3$ coverings, denoted $\mathcal{R}^{\mathbb{Z}_3 \times \mathbb{Z}_3}_2$ , is shown to consist of two disjoint irreducible components:

Isotropic component ( $\mathcal{R}_{iso}$ ): The subgroup defining the covering is isotropic with respect to the Weil pairing on the 3-torsion of the Jacobian $J_H[3]$ .
Non-isotropic component ( $\mathcal{R}_{ni}$ ): The subgroup is non-isotropic.
The authors analyze the geometry of these components separately, noting that the isotropic case presents additional technical difficulties due to the behavior of the restricted polarization.

B. Geometric Reconstruction Strategy (The "Top-Down" and "Bottom-Up" Approach)

To prove injectivity (Theorem 1.1), the authors develop a reconstruction algorithm that recovers the covering $f: \tilde{C} \to H$ from the polarized Prym variety $(P(f), \Theta_P)$ . The steps are:

Recovering the Galois Action: They prove that the action of the Galois group $\langle \sigma, \tau \rangle \cong \mathbb{Z}_3 \times \mathbb{Z}_3$ on the Prym variety can be uniquely recovered from the polarization $\Theta_P$ (Proposition 3.1).
Isotypical Decomposition: Using the group action, they decompose the Prym variety $P(f)$ into a product of four "small Pryms" (isotypical components), each corresponding to a quotient curve $C_\alpha = \tilde{C}/\langle \alpha \rangle$ where $\alpha \in \{\sigma, \tau, \sigma\tau, \sigma^2\tau\}$ .
Identifying Elliptic Curves:
- In the non-isotropic case, the restricted polarization on each small Prym uniquely determines three specific elliptic curves embedded within it.
- In the isotropic case, the small Pryms are not directly embedded; the authors must pass to a quotient by a carefully chosen 3-torsion point to recover the elliptic curves (Proposition 3.4).
Recovering Involutions and Quotients: The authors establish a correspondence between subsets of these elliptic curves and involutions in the automorphism group of the covering. Specifically, they identify 12 distinguished elliptic curves (quotients of $\tilde{C}$ by $S_3$ subgroups).
Reconstructing the Covering: By analyzing the "compatible 4-tuples" of these elliptic curves, they reconstruct the Jacobians of genus 4 curves ( $C_{ab}$ ). The Galois closure of the coverings from these genus 4 curves to elliptic curves uniquely determines the original curve $\tilde{C}$ and the covering map $f$ .

C. Generic Finiteness Analysis

For the general case (Theorem 1.2), the authors use dimension counting and fiber analysis:

They utilize the Torelli map and properties of isogenies.
For groups $G$ containing subgroups like $\mathbb{Z}_p \times \mathbb{Z}_p$ (for $p \ge 3$ ) or cyclic groups of order $\ge 6$ , they construct a map to the moduli of products of curves and show that the fibers are finite.
They leverage the fact that for cyclic groups of small order ( $<6$ ), the dimension of the Prym variety is too small relative to the moduli of the coverings, leading to positive-dimensional fibers.

3. Key Contributions and Results

Theorem 1.1: Global Injectivity for $\mathbb{Z}_3 \times \mathbb{Z}_3$

The Prym maps for both the isotropic and non-isotropic components are globally injective.

Isotropic Case: Maps to the moduli space of abelian varieties with polarization type $(1,1,1,1,1,1,3,3)$ .
Non-Isotropic Case: Maps to the moduli space with polarization type $(1,1,1,1,1,1,1,9)$ .
Significance: This proves a Prym-Torelli theorem for these specific coverings, meaning the covering is uniquely determined by its Prym variety. This extends previous results known for $\mathbb{Z}_2 \times \mathbb{Z}_2$ (Borówka & Orecchia) and cyclic coverings.

Theorem 1.2: Generic Finiteness for Arbitrary Abelian Groups

Let $G$ be a finite abelian group. The Prym map $P^G_2: \mathcal{R}^G_2 \to \mathcal{A}^G$ behaves as follows:

Positive Dimensional Fibers: If $G \cong \mathbb{Z}_2, \mathbb{Z}_3, \mathbb{Z}_4,$ or $\mathbb{Z}_5$ , the generic fiber is positive dimensional (the map is not generically finite).
Generically Finite: If $G$ is not isomorphic to a cyclic group of order less than 6 (i.e., $G$ contains a subgroup isomorphic to $\mathbb{Z}_6$ , $\mathbb{Z}_2 \times \mathbb{Z}_2$ , $\mathbb{Z}_3 \times \mathbb{Z}_3$ , $\mathbb{Z}_5 \times \mathbb{Z}_5$ , etc.), the Prym map is generically finite.

Note: The moduli space $\mathcal{R}^G_2$ is empty if the minimal number of generators of $G$ is $\ge 5$ .

Technical Lemmas and Propositions

Proposition 2.1: Establishes the decomposition of the moduli space into isotropic and non-isotropic components.
Lemma 2.5 & 2.10: Determine the precise polarization types of the Prym varieties and their isotypical components for both cases.
Proposition 3.8: Proves that the only "group-compatible" involutions on the Prym variety are those induced by the Galois group, a crucial step in the reconstruction.

4. Significance and Impact

Extension of Prym Theory: The paper significantly advances the understanding of Prym maps beyond the classical double coverings and cyclic coverings. It provides the first complete global injectivity result for non-cyclic abelian coverings of genus 2 curves with rank 2 ( $\mathbb{Z}_3 \times \mathbb{Z}_3$ ).
Methodological Innovation: The reconstruction technique, which utilizes the interplay between the Galois group action, the polarization, and the specific geometry of elliptic curves embedded in the Prym variety, offers a new template for studying other non-cyclic coverings.
Classification of Finiteness: The paper provides a comprehensive classification of when Prym maps for genus 2 coverings are generically finite. It resolves the boundary cases, showing that the "small" cyclic groups ( $<6$ ) are the only exceptions where the map fails to be generically finite.
Moduli Space Structure: The result that the non-isotropic moduli space $\mathcal{R}^{\mathbb{Z}_3 \times \mathbb{Z}_3}_{ni}$ can be embedded as an open dense subset of the moduli of $(1,9)$ -polarized abelian surfaces (Proposition 3.13) links the geometry of these coverings directly to the geometry of abelian surfaces.

In summary, this work establishes a robust Torelli-type theorem for $\mathbb{Z}_3 \times \mathbb{Z}_3$ coverings and clarifies the general landscape of Prym maps for abelian coverings of genus 2 curves, distinguishing clearly between cases where the map is finite and where it is not.