Fixed point locus of Moduli spaces of Sheaves on Toric DM stacks

This paper extends combinatorial descriptions of torsion-free toric sheaves to smooth toric Deligne-Mumford stacks of any dimension, explicitly characterizing the fixed point locus of their moduli spaces under torus actions via characteristic functions to facilitate the computation of topological invariants.

The Gist

Imagine you are trying to organize a massive, chaotic library. This library isn't just a building; it's a magical, multi-dimensional space where books (mathematical objects called "sheaves") can exist in strange, overlapping layers. Some books are whole and perfect, while others are torn or have missing pages.

The author of this paper, Promit Kundu, is trying to solve a specific puzzle: How do we find and count the "perfectly still" books in this library when the whole room is spinning?

Here is a breakdown of the paper's ideas using everyday analogies:

1. The Setting: A Spinning, Layered Library

The "library" in this paper is a Toric DM Stack.

• The "Toric" part: Imagine the library is built on a grid system, like a city with perfect streets and intersections. It has a lot of symmetry.
• The "Stack" part: This is the tricky bit. In a normal library, a book sits on a shelf. In this magical library, some shelves are "stacked" on top of each other in a way that creates hidden layers. It's like a book that is actually a set of three different books glued together, but you can only see one at a time depending on how you look at it.
• The "Spinning": The whole library is being rotated by a giant invisible hand (a mathematical "torus action"). Most books would fly off the shelves or blur into a mess as the library spins.

2. The Problem: Finding the "Still" Books

The author wants to study the Moduli Space. Think of this as a giant map or a catalog that lists every possible way you can arrange these books on the shelves.

When the library spins, most arrangements on the map would look different every second. But, there are special arrangements that look exactly the same even while the library spins. These are the Fixed Points.

• The Goal: The paper asks: "Can we describe these special, still arrangements without having to watch the whole library spin?"

3. The Solution: The "Characteristic Function" (The Fingerprint)

To find these still arrangements, the author invents a new way to describe the books called a Characteristic Function.

• The Analogy: Imagine every book in the library has a unique barcode made of numbers. In a normal library, the barcode just tells you the title. In this magical library, the barcode is much more detailed. It tells you exactly how the book is stacked, how many layers it has, and how it fits into the spinning grid.
• The "Box" Concept: The author breaks the library down into small rooms (open charts). In each room, the books are organized into "boxes" of data. The author proves that for a book to be "stable" (perfectly still), it must have exactly one box in each room. If it has two or more boxes in a room, it's unstable and will fall apart when the library spins.

4. The Gluing Formula: The Puzzle Pieces

The library is made of many overlapping rooms. To make a book that exists in the whole library, the data in Room A must match the data in Room B where they overlap.

• The Analogy: Imagine you are building a giant 3D puzzle. You have pieces for the corner, the edge, and the middle. The author creates a strict rule (a Gluing Formula) that says: "If you have a piece from the corner and a piece from the edge, here is exactly how they must snap together to make a valid whole."
• This rule ensures that the "barcode" (the characteristic function) is consistent everywhere.

5. The Big Discovery: The Decomposition

The paper's main result is a powerful simplification.

• Before: The map of all possible book arrangements is a giant, tangled, messy knot that is impossible to understand.
• After: The author shows that the "Still" part of this map (the fixed points) is actually just a collection of small, simple, separate islands.
• Each island corresponds to a specific type of barcode (a specific characteristic function).
• The Result: Instead of studying the giant, messy knot, mathematicians can now just study these small, simple islands one by one. The paper proves that the "Still" map is exactly the same as the sum of these simple islands.

6. Why This Matters (According to the Paper)

The author explains that by breaking the problem down into these small, combinatorial islands (the "barcodes"), it becomes much easier to calculate topological invariants.

• The Analogy: If you want to know the total weight of a giant, spinning pile of sand, it's hard. But if you realize the pile is just a collection of small, distinct buckets of sand, you can just weigh each bucket and add them up.
• The paper sets up the tools to do this "weighing" (calculating things like Euler characteristics) for these complex mathematical spaces.

Summary

In short, this paper takes a very complex, high-dimensional mathematical problem involving spinning, layered spaces and proves that the "still" parts of it can be completely understood by looking at simple, discrete patterns (barcodes). It turns a messy, continuous problem into a clean, countable puzzle.

Technical Summary

Technical Summary: Fixed Point Locus of Moduli Spaces of Sheaves on Toric DM Stacks

Problem Statement
The paper addresses the structural description of the moduli space of torsion-free sheaves on smooth projective toric Deligne–Mumford (DM) stacks. Specifically, it seeks to understand the fixed point locus of the natural torus action on the moduli space of modified Gieseker (or slope) stable torsion-free sheaves. A central challenge in this context is that standard stability conditions defined solely by an ample line bundle on the coarse moduli space fail to capture the stacky geometry. To resolve this, the paper employs the concept of a "generating sheaf" $\Xi$ to define a "modified Hilbert polynomial," ensuring the stability condition reflects the genuine stacky structure. The goal is to decompose the complex moduli space into a disjoint union of simpler components parametrized by combinatorial data.

Methodology
The approach extends combinatorial techniques previously developed for smooth toric varieties (by Klyachko, Perling, Kool, and others) to the setting of toric DM stacks. The methodology proceeds through several key technical steps:

1. Local Description via Stacky S-Families: The paper describes torsion-free toric sheaves locally on open charts $U_\sigma \cong [\mathbb{C}^d/G_\sigma]$ (corresponding to top cones of the fan) as "stacky S-families." These are collections of vector spaces equipped with fine gradings corresponding to the character groups of the torus $T$ and the finite stabilizer groups $G_\sigma$.
2. Box Decomposition and Gluing: A crucial component is the "box decomposition," where a sheaf on a chart is decomposed into a direct sum of subsheaves indexed by elements of a "box" $B_\sigma(T)$. The paper establishes a rigorous Gluing Formula (Theorem 3.8) that dictates how these local stacky S-families must match across intersections of charts to form a global sheaf. This involves analyzing the fiber product of open substacks and imposing compatibility equalities on the representations of the intersection groups.
3. Characteristic Functions: To parametrize the fixed points, the author introduces characteristic functions ($\vec{\chi}$). These functions encode the dimensions of the weight spaces associated with the torus action for each chart, subject to the gluing constraints. For stable sheaves, the paper proves (Theorem 3.11) that indecomposability forces the existence of a unique "box element" per chart, simplifying the characteristic function to a specific form.
4. Matching Stability Conditions: The paper constructs moduli spaces of sheaves with fixed characteristic functions using Geometric Invariant Theory (GIT). A critical technical contribution is the proof that GIT stability (with respect to a specific equivariant line bundle) is equivalent to modified Gieseker (or slope) stability for these toric sheaves (Theorems 4.2 and 4.3). This matching allows the translation of geometric stability conditions into combinatorial GIT problems.
5. Fixed Point Decomposition: By showing that the torus action lifts to the moduli space and that the fixed points correspond precisely to toric sheaves with specific equivariant structures, the author establishes an isomorphism between the fixed point locus and a disjoint union of moduli spaces of sheaves with fixed characteristic functions.

Key Contributions and Results

• Combinatorial Description of Fixed Points: The primary result (Theorem 1.1 and Theorem 4.6) provides a canonical isomorphism between the fixed point locus $(M^s_{P_\Xi})^T$ of the moduli space of modified stable torsion-free sheaves and a disjoint union of moduli spaces $M^s_{\vec{\chi}}$ parametrized by framed characteristic functions:
  $$(M^s_{P_\Xi})^T \cong \bigsqcup_{\vec{\chi} \in (\chi_{P_\Xi})_{fr}} M^s_{\vec{\chi}}$$
  This reduces the study of the moduli space to combinatorial data.
• Extension to Reflexive Sheaves: A parallel result (Theorem 1.2 and Theorem 4.7) is established for the moduli space of modified slope-stable reflexive sheaves, which are a crucial subclass often relevant to instanton moduli spaces.
• Gluing Formula for DM Stacks: The paper derives a general gluing formula (Theorem 3.8) for torsion-free toric sheaves on arbitrary smooth toric DM stacks, handling the complexities of non-primitive torus actions and finite stabilizer groups.
• Explicit Invariants: The techniques are applied to compute generating functions for topological invariants. Specifically, Example 3 and Example 4 provide explicit formulas for the generating functions of modified Euler characteristics for rank 1 sheaves on toric surface orbifolds and weighted projective planes, utilizing torus localization.

Significance
The paper claims to generalize the work of Klyachko, Perling, and Kool from smooth toric varieties to the broader and more complex setting of smooth toric DM stacks. By incorporating the generating sheaf $\Xi$, the work ensures that the stability conditions and resulting moduli spaces are sensitive to the stacky structure, which is lost if one only considers the coarse moduli space.

The significance lies in providing a concrete, combinatorial framework to study these moduli spaces. By decomposing the fixed point locus into components parametrized by characteristic functions, the paper enables the computation of topological invariants (such as Betti numbers and Euler characteristics) via torus localization techniques. The author notes that this framework builds upon and extends previous results on weighted projective planes and stacky Hirzebruch orbifolds, offering a unified approach for arbitrary smooth projective toric DM stacks. The work serves as a foundational step for future investigations into higher-dimensional stacky toric threefolds and wall-crossing phenomena.