🔢 mathematics

Birational Classification of Orbifold Compactified Jacobians

This paper establishes an equivariant orbifold birational classification for toroidal compactifications of algebraic tori and semiabelian schemes by reducing the problem to a combinatorial search for minimal orbifold toroidal compactifications in logarithmic geometry, thereby generalizing and providing a geometric interpretation of recent results by Schmitt.

Original authors: Jeremy Feusi, Sam Molcho

Published 2026-05-06

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jeremy Feusi, Sam Molcho

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 an architect trying to organize a vast, chaotic city. In this city, there are different types of buildings: some are simple houses (algebraic varieties), and others are complex structures with hidden rooms, secret passages, and special rules for who can enter (orbifolds or "stacks").

The paper by Jeremy Feusi and Sam Molcho is about a specific problem: How do we decide if two different versions of this city are essentially the "same" from a structural perspective, even if they look different on the surface?

Here is a breakdown of their work using simple analogies:

1. The Problem: Are Two Cities the Same?

In traditional geometry, if you can turn one building into another by cutting and pasting (without tearing), they are considered "birationally equivalent." But in the world of orbifolds (these complex, rule-heavy structures), the old rules don't work.

The authors use a new definition of "sameness" proposed by Kresch and Tschinkel. Imagine two cities are "equivalent" only if you can build a third, temporary bridge city that connects to both of them via sturdy, reversible bridges. If you can do this, the two cities are "birationally equivalent."

The challenge is that for these complex orbifold cities, there are infinitely many ways to build them, and it's hard to tell which ones are truly unique.

2. The Solution: The "Logarithmic" Blueprint

The authors realize that to solve this, they need to stop looking at the buildings as they are and start looking at their "logarithmic" blueprints.

Think of a "log scheme" as a building that comes with a detailed instruction manual attached to its walls. This manual tells you exactly how the building interacts with its surroundings (the "boundary").

The Goal: They want to find the minimal version of any city. This is the "smallest" possible version that still contains all the essential information.
The Analogy: Imagine you have a messy, overgrown garden (a complex orbifold). You want to find the "minimal" garden that keeps the same flowers and paths but removes all the extra, unnecessary bushes. The authors prove that for certain types of gardens (specifically those related to Jacobians of curves and tori), there is always one unique, perfect "minimal" version.

3. The Magic Trick: Turning Geometry into Puzzles

The most exciting part of the paper is how they solve the problem. They translate the complex geometry of these cities into a purely combinatorial puzzle (a puzzle made of shapes and numbers).

The Tropical Correspondence: They use a tool called "tropical geometry." Imagine taking a 3D sculpture and projecting its shadow onto a flat wall. The shadow loses some detail but keeps the essential shape.
The Shadow: In their case, the "shadow" is a collection of cones and lattices (like a 3D grid made of paper cones).
The Discovery: They prove that there is a perfect one-to-one match between the complex geometric cities and these simple cone puzzles. If you can solve the puzzle, you know exactly what the city looks like.

4. The "Sublattice Coloring" (The Final Answer)

For the specific case of Toric Orbifolds (a type of city built around a central torus shape), they solve the puzzle completely.

They find that every unique city corresponds to a specific way of coloring a grid.

Imagine a grid of points.
You are allowed to pick a smaller, denser grid inside it (a "sublattice").
You "color" the points based on which grid they belong to.
The Result: Every unique way of coloring this grid corresponds to a unique type of orbifold city. This generalizes a recent result by a mathematician named Schmitt, but the authors explain why it works using their "logarithmic" framework.

5. What They Actually Solved (and What They Didn't)

The paper makes a clear distinction between what is proven and what is a guess:

Proven: They successfully classified these "cities" when the underlying structure is a Torus (like a donut shape) or a Jacobian (related to families of curves with nodes, like a chain of circles). They showed that for these, there is a unique minimal version, and it can be found by solving the cone puzzle.
The Conjecture (The Open Question): They suspect this method works for all semi-abelian schemes (a broader class of structures). However, to prove it, they need to assume a specific property (the "Néron mapping property") holds true for all these structures. They state this as a conjecture. If this conjecture is true, their classification works for everything; if not, their proof only works for the specific cases they checked.

Summary

In short, Feusi and Molcho took a very hard problem about classifying complex geometric shapes. They invented a new way to look at these shapes (using "logarithmic" blueprints), turned the problem into a puzzle of cones and grids, and solved the puzzle for specific, important types of shapes. They showed that every complex shape has a unique "minimal" core, and they can identify that core by looking at how the shape's grid is colored.

They didn't invent a new building material or predict how this helps build real houses; they simply provided a new, clearer map for mathematicians to navigate the landscape of these abstract geometric worlds.

Technical Summary: Birational Classification of Orbifold Compactified Jacobians

Problem Statement
The paper addresses the equivariant orbifold birational classification problem for families of toroidal compactifications of a group $G$ over a toroidal base $S$ . Specifically, it focuses on cases where $G$ is an algebraic torus, a semiabelian scheme, or the Jacobian of a family of nodal curves. The authors work within the framework of Deligne-Mumford (DM) stacks, adopting the definition of birational equivalence proposed by Kresch and Tschinkel, which requires the existence of a correspondence with proper, birational, and representable morphisms. This definition distinguishes orbifold birational classes from those of algebraic varieties, as it accounts for the stacky structure (isotropy groups) and prevents the trivialization of orbifold classes into single variety classes.

The central difficulty lies in the fact that while the classical birational classification of complete toric varieties is trivial (determined solely by dimension), the orbifold version yields infinitely many distinct classes. The paper aims to classify these classes for semiabelian schemes and Jacobians, generalizing recent results by Schmitt on toric orbifolds.

Methodology
The authors employ the language of logarithmic geometry to translate the geometric classification problem into a combinatorial one. The methodology proceeds through several key stages:

Logarithmic Compactifications and Artin Fans: The authors define log compactifications of a semiabelian scheme $G$ as proper, log smooth, $G$ -equivariant morphisms $X \to S$ containing $G$ as a dense open substack. They establish a correspondence between these geometric objects and Artin fans (stacks over the category of rational polyhedral cones). Specifically, they show that the category of log compactifications is equivalent to the category of "combinatorial partial log compactifications" via a tropicalization functor.
Tropical Correspondence: A "tropicalization" functor $\Sigma_?$ is constructed, mapping log compactifications to "tropical compactifications" (subdivisions of tropical semiabelian varieties). Conversely, a "geometric realization" functor $R_?$ reconstructs the geometric object from the combinatorial data. The authors prove that this correspondence is an equivalence of categories, reducing the geometric problem to the classification of tropical compactifications.
Generalized Objects and Minimality: To achieve a complete classification, the authors introduce generalized log compactifications (allowing for certain non-representable morphisms in the combinatorial limit) and minimal objects. They demonstrate that every birational equivalence class contains a unique minimal object. This minimal object serves as the canonical representative of the class.
Combinatorial Solution: The problem of identifying minimal objects is solved combinatorially. The authors define "stacky fans" and "minimal fans" (subsets of lattices satisfying specific intersection and invariance properties). They prove that minimal generalized tropical compactifications correspond bijectively to these minimal fans.
Néron Mapping Property: A crucial technical step involves the use of the Néron mapping property for logarithmic Jacobians (established by Holmes et al.). This property ensures that any log compactification of a Jacobian admits a unique equivariant morphism to the logarithmic Jacobian, allowing the authors to treat Jacobians within the same framework as algebraic tori.

Key Contributions and Results

Generalization of Schmitt's Results: The paper extends Schmitt's classification of toric orbifolds (which used "sublattice colorings") to the broader context of semiabelian schemes and Jacobians. It provides a geometric interpretation of Schmitt's sublattice colorings as the fans of minimal log compactifications.
Classification of Compactified Jacobians: The authors provide a complete birational classification for compactified Jacobians of log smooth curves. They establish a bijection between the set of birational equivalence classes of compactified Jacobians and the set of minimal generalized tropical compactifications of the tropical Jacobian.
Minimal Log Compactifications: The paper proves that for any semiabelian scheme $G$ (under specific conditions), the category of log compactifications contains a unique minimal object, denoted $G^{\log}$ . While any log compactification maps to $G^{\log}$ , the map is only a birational equivalence if it is representable. This clarifies the structure of the birational classes: they are in one-to-one correspondence with the minimal generalized log compactifications.
Combinatorial Description: The authors provide an explicit combinatorial description of these minimal classes. For a torus $T$ , the classes correspond to "sublattice colorings" (generalized to the semiabelian case). For Jacobians, they correspond to subsets of the tropical Jacobian defined by complete stacky fans.
Reduction to Conjecture: The general semiabelian case is reduced to an open conjecture regarding the Néron mapping property for log abelian varieties. The authors state that if this conjecture holds, their results generalize to arbitrary families of semiabelian varieties admitting a compactification by a log abelian variety.

Significance and Claims
The paper claims to place the recent combinatorial results of Schmitt [Sch24] into a natural conceptual framework within logarithmic geometry. By utilizing the language of log schemes and Artin fans, the authors geometrically interpret sublattice colorings as the data defining minimal log compactifications.

The significance of the work lies in its ability to solve a "strictly harder" classification problem (orbifold birational classification) by reducing it to a combinatorial problem involving tropical geometry and lattice subdivisions. The authors emphasize that their notion of birational equivalence is relative to the base $S$ , which is a necessary refinement for families of varieties.

The paper does not claim to solve the general semiabelian case unconditionally but rather reduces it to a well-motivated conjecture (Conjecture 1) concerning the Néron mapping property. The results are presented as a generalization and geometric interpretation of existing work, providing a unified framework for understanding the birational geometry of toric varieties, semiabelian schemes, and Jacobians in the orbifold setting.