🔢 mathematics

Shellings of Unbounded Polyhedra

This paper investigates the shellability of unbounded polyhedra through suitable compactifications to derive results for tropical hypersurfaces and reveal a connection between polyhedral duality and discrete Morse theory, demonstrating that the tight span of a regular subdivision is collapsible but not necessarily shellable.

Original authors: George Balla, Michael Joswig, Lena Weis

Published 2026-07-14

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: George Balla, Michael Joswig, Lena Weis

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 have a giant, invisible net made of rubber bands and strings, stretching out forever into the sky. This net represents a mathematical shape called an unbounded polyhedron. It's like a city of rooms that goes on forever in some directions, with no roof and no walls to stop you from walking off into infinity.

Mathematicians love to study these shapes by breaking them down into smaller, manageable pieces, like taking apart a Lego castle brick by brick. This process is called shellability. Think of it like peeling an onion or stacking pancakes: if you can remove the layers one by one without tearing the rest of the structure apart, and if each layer you remove is a nice, simple shape, then the whole thing is "shellable." This is a superpower because it tells us the shape isn't too weird or twisted; it's "simple enough" to understand.

The Problem: The Infinite Mess

Here's the snag: you can't really peel an onion that never ends. The standard rules for shellability only work on shapes that are finite and closed (like a regular ball or a cube). But our "polyhedron city" stretches to infinity. If you try to apply the rules directly, the math breaks because the shape isn't compact (it doesn't fit in a box).

The Solution: The Truncation Trick

The authors, George, Michael, and Lena, came up with a clever way to handle this. They decided to pretend the infinite city has a giant, invisible wall around it.

The Truncation: Imagine taking a giant, smooth, convex ball (like a giant beach ball) and placing it over the city so that it covers all the "starting points" (the vertices) but cuts off the parts stretching to infinity. Where the ball slices through the infinite streets, it creates new, flat walls. Suddenly, the infinite city becomes a finite, closed shape—a truncated polyhedron.
The Result: They proved that if you do this, you can find a perfect order to peel the layers off this new, finite shape. It's like finding a specific path through the city that lets you visit every room and exit without getting stuck.

They also tried a second method called Tropical Toric Compactification. This is a bit more like folding the infinite map onto a special, curved surface (a "tropical toric variety") where the edges meet up nicely. They found that this method works too! You can still peel the layers off in a perfect order.

The Big Application: Tropical Hypersurfaces

Why does this matter? The authors use this trick to solve a puzzle in tropical geometry. Think of tropical geometry as a weird, sunny version of math where "addition" means taking the minimum and "multiplication" means adding. In this world, equations draw shapes called tropical hypersurfaces. These are like the ridges on a mountain range where the ground is lowest in two different directions at once.

These tropical shapes are actually the "shadows" or "skeletons" of the infinite polyhedra the authors just studied. Because they figured out how to shell the infinite polyhedra (by truncating them or compactifying them), they proved that these tropical shapes are also shellable.

What this means: It turns out that the "homotopy type" (the basic shape) of these tropical hypersurfaces is just a bunch of spheres stuck together at a single point, like a bouquet of flowers. The number of flowers depends on how many "holes" or "inner points" are in the underlying grid.

The Twist: Not Everything is Shellable

Here is where the story gets spicy. The authors asked: "What about the tight span?"

The tight span is like the "dual" or the "shadow" of the tropical shape. It's a different way of looking at the same data. You might think, "If the original shape is shellable, its shadow must be too!"

The paper explicitly rules this out.
The authors found a specific example (a tropical plane curve with a genus of two, which is like a figure-eight shape) where the tight span is not shellable. You cannot peel the layers off this specific shape in a perfect order without breaking it. So, while the original tropical shape is nice and tidy, its dual twin can be a messy, non-shellable knot.

The Silver Lining: Collapsibility

But don't worry! Even though the tight span isn't shellable, it's not a total disaster. The authors proved that these tight spans are always collapsible.

Think of collapsibility as a slightly weaker version of shellability. If shellability is like carefully peeling an onion layer by layer, collapsibility is like a magic trick where the shape folds in on itself until it becomes a single point, without tearing. It's like a pop-up card that folds flat.

The authors showed that no matter how messy the tight span looks, it can always be folded down to a single point. This means it's contractible (it can shrink to a dot), which is a very strong property. They proved this using a tool called discrete Morse theory, which is like a game of matching pairs of cells to see which ones can be removed to simplify the shape.

Summary of the Findings

Proven: The authors proved that if you take an infinite polyhedron and cut it off with a wall (truncation) or fold it onto a special surface (compactification), you can always find a perfect order to peel it apart (it is shellable).
Proven: This means tropical hypersurfaces (the ridges in tropical geometry) are also shellable.
Proven: The "tight spans" (the dual shapes) are always collapsible (they can fold into a point).
Proven: However, the authors explicitly showed that tight spans are not always shellable. There are specific examples where the perfect peeling order doesn't exist.
Unknown: The paper ends with a question: Is the tight span of every tropical polytope shellable? They don't know the answer yet. They found one that isn't, but they haven't proven that none of them are, or that all of them are. It's an open mystery.

So, in the world of infinite shapes, the authors found a way to tame the wild edges, proved that the main shapes are tidy, but discovered that their shadows can be a bit messy—though never too messy to fold into a single point.

Technical Summary: Shellings of Unbounded Polyhedra

Problem Statement
The paper addresses the shellability of the boundary complexes of unbounded polyhedra. While shellability is a well-established concept for finite regular cell complexes (specifically bounded polytopes), extending this notion to unbounded polyhedra presents a technical obstacle: the boundary of an unbounded polyhedron is not compact and thus does not form a finite regular cell complex in the standard sense. The authors aim to resolve this by studying suitable compactifications of unbounded polyhedra to derive shellability results. These results are then applied to tropical geometry, specifically to tropical hypersurfaces and the tight spans of regular subdivisions. A secondary problem addressed is the topological nature of tight spans, which are known to be contractible but whose shellability properties were previously unclear.

Methodology
The authors employ a combination of polyhedral geometry, combinatorial topology, and discrete Morse theory.

Compactification Strategies: To handle unboundedness, the paper investigates two specific compactification methods for a pointed, unbounded polyhedron $P$ :
- Truncation ( $P'$ ): Intersecting $P$ with a half-space $A$ that contains all vertices of $P$ in its interior. This creates a bounded polytope $P'$ with an additional "bounding facet."
- Tropical Toric Compactification ( $P^\diamond$ ): Following Kastner, Shaw, and Winz, this involves embedding $P$ into a tropical toric variety associated with its recession cone. This adds "faces at infinity" corresponding to the faces of the dual of the recession cone.
Line Shelling Extension: The authors extend the classical Bruggesser–Mani line shelling method (originally for bounded polytopes) to these compactified structures. They utilize generic affine lines to induce orderings on facets. For truncations, they show that a line can be oriented such that the bounding facet appears last in the shelling order. For tropical toric compactifications, they prove shellability by inductively combining shellings of the truncation of the polyhedron and shellings of the star of the dual cone.
Duality and Tropical Geometry: The paper leverages the duality between unbounded polyhedra (specifically "domes" $D(A, \omega)$ ) and regular subdivisions (normal complexes $NC(A, \omega)$ ). The tropical hypersurface is identified as the $(d-1)$ -skeleton of the normal complex. By establishing shellability for the compactified polyhedra, they induce shellability for the compactified tropical hypersurfaces.
Discrete Morse Theory: To address the collapsibility of tight spans (which are not necessarily shellable), the authors utilize Chari's adaptation of Forman's discrete Morse theory. They construct acyclic matchings on the Hasse diagram of the cell poset to demonstrate collapsibility.

Key Contributions and Results

Shellability of Compactified Unbounded Polyhedra:
- Proposition 4: The boundary of the truncation $P'$ of a pointed unbounded polyhedron $P$ is shellable, with the bounding facet appearing last in the shelling order. This induces shellings on the truncations of the unbounded facets of $P$ .
- Theorem 12 & Corollary 13: The tropical toric compactification $P^\diamond$ of a polyhedron $P$ is shellable. The shelling order is constructed by combining a shelling of the truncation of the polyhedron and a shelling of the star of the dual recession cone.
- Corollary 14: The tropical toric compactification of the $k$ -skeleton of $P$ is also shellable.
Application to Tropical Hypersurfaces:
- Theorem 24: The truncated tropical hypersurface $T(F) \cap B$ in $\mathbb{R}^d$ is a shellable regular cell complex.
- Theorem 25: The tropical toric compactification of a tropical hypersurface is a shellable regular cell complex.
- Corollary 26: As a consequence of shellability, the homotopy type of a tropical hypersurface is a wedge of $(d-1)$ -spheres. The number of spheres corresponds to the number of interior vertices of the dual regular subdivision. This strengthens a previous observation by Mikhalkin and Rau.
- Corollary 30: Tropical hyperplane arrangements in the tropical projective torus are shellable.
Collapsibility vs. Shellability of Tight Spans:
- Example 22: The authors provide an explicit counterexample showing that the tight span of a regular subdivision (specifically related to a genus-two tropical plane curve) is not necessarily shellable.
- Theorem 35: Despite the lack of shellability, the tight span of any regular subdivision is collapsible. The proof constructs an acyclic matching on the Hasse diagram of the cell complex, showing it has exactly one critical cell. This implies the tight span is contractible, reinforcing known results but providing a stronger topological characterization via collapsibility.
- Theorem 33: For covector decompositions induced by tropical hyperplane arrangements, ordering maximal cells by their coarse types lexicographically induces a shelling of both the truncation and the tropical toric compactification.

Significance and Claims
The paper claims to resolve the technical difficulty of defining shellability for unbounded polyhedral complexes by introducing and analyzing two robust compactification schemes. The primary significance lies in transferring these combinatorial results to tropical geometry, thereby establishing that tropical hypersurfaces and hyperplane arrangements possess the strong structural property of shellability (in their compactified forms).

Furthermore, the paper clarifies the topological status of tight spans. While it confirms they are not generally shellable, it proves they are always collapsible. This distinction is significant because collapsibility implies contractibility but is a stronger condition than mere contractibility, offering a more refined understanding of the combinatorial topology of these objects. The work relies on the interplay between the duality of polyhedral complexes and their shellability, translating these geometric insights into the language of discrete Morse theory.

The authors remain modest regarding the scope of their findings, noting that while they prove collapsibility for tight spans, the general question of whether the covector decomposition of every tropical polytope is shellable remains open (Question 40). They also acknowledge that the relationship between their results and the "tropical shellability" of fans defined by Amini and Piquerez requires further investigation.