Locally Trivial Deformations of Toric Varieties

This paper introduces a combinatorial deformation functor $\mathrm{Def}_\Sigma$ based on Čech cochains that is isomorphic to the locally trivial deformation functor of toric varieties under specific conditions, enabling new criteria for unobstructedness, explicit obstruction formulas, and a classification of unobstructed iterated $\mathbb{P}^1$-bundle threefolds.

The Gist

Imagine you have a beautiful, intricate sculpture made of geometric blocks. In the world of mathematics, this sculpture is called a Toric Variety. It's built from a specific blueprint called a Fan, which is just a collection of cones (like ice cream cones) arranged in a specific pattern.

Now, imagine you want to wiggle this sculpture slightly. Maybe you stretch one part, shrink another, or twist a corner. In math, this is called a deformation. The big question mathematicians ask is: Can I wiggle this sculpture in any direction I want without it breaking, or are there hidden "glue" spots that prevent me from moving certain parts?

If a sculpture can be wiggled freely in all directions, it has an unobstructed deformation space. If it gets stuck or breaks when you try to move it in two directions at once, it is obstructed.

This paper by Nathan Ilten and Sharon Robins is like a new instruction manual for figuring out exactly how these geometric sculptures can be wiggled. Here is the breakdown in simple terms:

1. The Problem: It's Too Complicated to Look at the Whole Sculpture

Usually, to see if a sculpture can be wiggled, you have to look at the whole thing at once. This is like trying to fix a giant, complex clock by staring at the entire machine. It's overwhelming and hard to calculate.

The authors realized that because these sculptures are built from a rigid, combinatorial blueprint (the Fan), you don't need to look at the whole clock. You can look at the individual gears and springs (the mathematical building blocks) and figure out how they interact.

2. The Solution: A "Combinatorial Deformation Functor"

The authors created a new tool they call a Combinatorial Deformation Functor. Think of this as a Lego instruction sheet that tells you exactly which bricks can be swapped out and which ones are locked together.

• The Old Way: You try to move the whole sculpture and hope it doesn't break.
• The New Way: You look at the blueprint (the Fan). You check specific little clusters of cones (called simplicial complexes). If these little clusters are "connected" in a certain way, the sculpture is flexible. If they are "broken" or disconnected, the sculpture is rigid or obstructed.

They turned a massive, scary geometry problem into a puzzle you can solve with logic and counting.

3. The "Cup Product" and Higher-Order Obstructions

Imagine you have two different ways to wiggle your sculpture:

1. Wiggle it North.
2. Wiggle it East.

If you try to wiggle it North-East (a combination of both), does it work?

• The Cup Product: This is a mathematical formula that checks if "North" and "East" play nice together. If they clash, the sculpture breaks.
• The New Discovery: The authors found that sometimes, "North" and "East" play nice, but when you add a third direction, "South," suddenly everything breaks! This is called a higher-order obstruction.

They developed a new, complex formula (like a super-advanced recipe) to predict exactly when these "triple clashes" happen. They found that sometimes the sculpture breaks not because of a simple clash, but because of a complicated interaction between three or more moves.

4. The Surprising Results

Using their new Lego-instruction manual, they tested many different sculptures and found some shocking things:

• Small is Safe: If the sculpture is very simple (low "Picard rank," which is like having very few types of blocks), it is always flexible. You can wiggle it however you want.
• The "Twisted" Surprise: They found specific 3D sculptures (built by stacking layers of 2D shapes) that look smooth and perfect but have hidden "kinks."
  • Some of these sculptures have two different "universes" of ways to wiggle them. One universe is smooth, but the other is "crumpled" (mathematically called non-reduced). It's like having a sculpture that can be stretched smoothly, or squashed into a weird, sticky shape that doesn't want to let go.
  • They found cases where the "crumpled" universe is infinitely larger than the smooth one.

5. Why This Matters

Before this paper, mathematicians thought these geometric sculptures were either perfectly flexible or had simple, predictable problems. This paper shows that the reality is much wilder.

• It breaks "Murphy's Law": In math, "Murphy's Law" suggests that deformation spaces can be as ugly and broken as possible. These authors showed that while these sculptures can be obstructed, they aren't arbitrarily ugly. They have a specific, predictable structure.
• New Tools for the Future: By giving us a way to calculate these "wiggles" using simple counting and blueprints, they allow other mathematicians to design new shapes, study mirror symmetry (a concept in physics where two different shapes describe the same universe), and understand the boundaries of geometric possibility.

The Bottom Line

Ilten and Robins took a problem that was like trying to untangle a giant knot of spaghetti and turned it into a game of Tetris. They showed that by looking at the individual blocks and how they fit together, you can predict exactly how the whole structure will behave, revealing hidden complexities that no one knew existed before.

Technical Summary

1. Problem Statement and Motivation

The paper addresses the deformation theory of toric varieties, specifically focusing on locally trivial deformations. While the deformation theory of smooth varieties is generally controlled by the cohomology of the tangent sheaf $H^1(X, T_X)$ (infinitesimal deformations) and $H^2(X, T_X)$ (obstructions), explicitly computing the deformation functor $\text{Def}_X$ for specific toric varieties has been challenging.

• The Gap: Previous work (e.g., by Ilten, Vollmert, Mavlyutov) provided combinatorial descriptions of first-order deformations and homogeneous deformations. However, combining these into a full deformation space often encounters obstructions.
• The Challenge: Unlike smooth Calabi-Yau or Fano varieties (which are often unobstructed), smooth complete toric varieties can be obstructed. Yet, they do not satisfy "Murphy's Law" (arbitrarily bad singularities), suggesting their deformation spaces have a specific, yet unknown, structure.
• Goal: To provide a purely combinatorial description of the locally trivial deformation functor for $Q$-factorial complete toric varieties, enabling the explicit computation of their deformation spaces (hulls) and obstruction theories.

2. Methodology

The authors develop a framework that translates the geometric deformation problem into a combinatorial one using simplicial complexes and Čech cohomology.

A. Combinatorial Description of Cohomology

The paper relies on the isomorphism between the cohomology of the tangent sheaf and the reduced cohomology of specific simplicial complexes associated with the fan $\Sigma$:
$$H^k(X_\Sigma, T_{X_\Sigma}) \cong \bigoplus_{\rho, u} \tilde{H}^{k-1}(V_{\rho, u}, K)$$
where:

• $\rho$ ranges over rays of the fan.
• $u$ ranges over characters of the torus.
• $V_{\rho, u}$ is a simplicial complex derived from the fan geometry and the character $u$.

B. The Combinatorial Deformation Functor ($\text{Def}_\Sigma$)

The core innovation is the construction of a functor $\text{Def}_\Sigma$ defined entirely via Čech zero-cochains on the complexes $V_{\rho, u}$.

1. Lie Algebra Structure: The authors define a Lie bracket on the direct sum of sheaves $\bigoplus O(D_\rho)$ (related to the Cox torsor), which maps surjectively to the tangent sheaf $T_X$.
2. Functor Definition: For a local Artinian algebra $A$, $\text{Def}_\Sigma(A)$ consists of elements $\alpha$ in the space of zero-cochains such that a specific "obstruction map" $o_\Sigma(\alpha) = 0$, modulo an equivalence relation induced by the Lie algebra action.
3. Main Theorem (Isomorphism): Under the hypotheses that $X$ is $Q$-factorial, has no torus factors, and $H^1(X, \mathcal{O}_X) = H^2(X, \mathcal{O}_X) = 0$ (e.g., smooth complete toric varieties), the combinatorial functor $\text{Def}_\Sigma$ is isomorphic to the geometric functor of locally trivial deformations $\text{Def}'_X$.

C. Algorithmic Hull Computation

The authors adapt Schlessinger's theory to construct the hull (the universal deformation space) of $\text{Def}_\Sigma$ algorithmically:

• Deformation Equation: They define a "combinatorial deformation equation" involving a map $\psi$ that lifts cocycles.
• Iterative Solving: By iteratively solving this equation to higher orders, they compute the defining ideal of the hull.
• Higher Order Obstructions: They derive explicit formulas for obstructions of arbitrary order using the Baker-Campbell-Hausdorff (BCH) product. This generalizes previous cup product formulas, showing that higher-order obstructions depend not just on first-order data but on higher-order perturbation data.

D. Simplification Techniques

To make computations feasible, the paper introduces:

• Fan Reduction: A method to remove certain maximal cones from the fan $\Sigma$ without changing the deformation hull, provided the reduced fan still "covers" the relevant combinatorial data (set $\Gamma$).
• Cocycle Property: A technique to reduce the number of pairwise cone intersections that need to be checked when computing obstructions, based on the connectivity of the fan's star neighborhoods.

3. Key Contributions and Results

A. Theoretical Isomorphism

Theorem 1.2.1: Establishes that for smooth complete toric varieties (and more generally, those smooth in codimension 2 and $Q$-factorial in codimension 3), the geometric deformation functor is isomorphic to the combinatorial functor $\text{Def}_\Sigma$. This allows the use of purely combinatorial tools to solve geometric deformation problems.

B. Unobstructedness Criteria

Theorem 1.2.2: Provides a sufficient condition for a toric variety to have an unobstructed deformation space. If certain cohomology groups $H^1(V_{\rho, u}, K)$ vanish for a specific set of ray-character pairs (generated by the first-order deformations), the variety is unobstructed.

• Significance: This allows for the verification of unobstructedness in cases where degree arguments alone fail (e.g., Example 5.5.2).

C. Classification of Picard Rank 2 and 3 Varieties

• Picard Rank 1 & 2: The authors prove that all smooth complete toric varieties of Picard rank 1 or 2 have unobstructed deformations. They also characterize exactly when such varieties are rigid.
• Picard Rank 3 (Iterated $\mathbb{P}^1$-bundles): They classify the obstruction behavior for toric threefolds that are iterated $\mathbb{P}^1$-bundles over Hirzebruch surfaces ($X = \mathbb{P}(\mathcal{O}_{F_e} \oplus \mathcal{O}_{F_e}(aF+bH))$).
  • They identify specific regions in the parameter space $(e, a, b)$ where obstructions occur.
  • They determine the minimal degree of obstructions: some cases have quadratic obstructions, while others have cubic obstructions.

D. Discovery of New Phenomena

By explicitly computing hulls for specific examples, the authors discover deformation spaces with properties previously unobserved for smooth toric varieties:

1. Generically Non-Reduced Components: The hull can have a component that is non-reduced (multiplicity > 1) generically.
2. Irreducible but Singular: The hull can be irreducible but possess a singularity at the origin.
3. Arbitrarily Large Dimension Differences: The hull can consist of two irreducible components with an arbitrarily large difference in their dimensions.
4. Non-Quadric Obstructions: They disprove the conjecture (Question 1.4 in [IT20]) that deformation spaces of smooth toric varieties are always cut out by quadrics. They exhibit examples where the defining equations are cubic.

4. Significance and Impact

• Computational Power: The paper provides a concrete, algorithmic pipeline (implemented in Macaulay2) to compute deformation spaces for toric varieties, moving beyond theoretical existence to explicit calculation.
• Structural Insight: It reveals that the deformation theory of toric varieties is richer and more complex than previously thought, exhibiting phenomena like non-reducedness and high-degree obstructions.
• Connection to Cox Rings: The methodology highlights the deep connection between the deformation of a toric variety and the invariant deformations of its Cox torsor, offering a new perspective on the geometry of toric varieties.
• Applications: The results have implications for mirror symmetry (deformations of Calabi-Yau hypersurfaces), the classification of Fano varieties, and the study of singularities in moduli spaces.

In summary, Ilten and Robins successfully bridge the gap between the combinatorial data of a fan and the complex geometric structure of its deformation space, providing both a theoretical isomorphism and practical tools to uncover the intricate nature of toric deformations.