🔢 mathematics

Algebraizability of Vector Bundles over Real Algebraic Varieties

This paper utilizes motivic homotopy theory to establish that while the algebraicity of Stiefel-Whitney classes is sufficient for the algebraizability of topological vector bundles over affine smooth real algebraic varieties of dimension at most three, a new obstruction involving the first Pontryagin and fourth Stiefel-Whitney classes arises in the four-dimensional compact case.

Original authors: Hanqi Wang

Published 2026-07-10

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hanqi Wang

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 build a house. You have two blueprints: one is a Topological Blueprint, drawn with flexible, stretchy rubber lines that can be bent and twisted without tearing. The other is an Algebraic Blueprint, drawn with rigid, mathematical rules where every line must follow a strict equation.

The big question this paper asks is: If you have a house built from the flexible rubber blueprint, can you always find a rigid algebraic blueprint that looks exactly the same?

In the world of math, these "houses" are called vector bundles (think of them as layers of fabric or fields of arrows covering a shape), and the "shapes" they sit on are real algebraic varieties (geometric shapes defined by equations over real numbers).

The Golden Rule for Small Shapes (Dimensions 1, 2, and 3)

The authors, led by Hanqi Wang, discovered a beautiful rule for shapes that are 3 dimensions or smaller (like a solid ball, a donut, or a twisted tube).

They found that for these small shapes, you don't need to check the whole complex blueprint. You only need to check two specific "tags" attached to the house, called Stiefel-Whitney classes (let's call them Tag 1 and Tag 2).

The Finding: If Tag 1 and Tag 2 are "algebraic" (meaning they follow the rigid math rules), then the entire house can be rebuilt using the rigid algebraic blueprint.
The Certainty: This is a proven fact for any smooth, 3-dimensional shape. If the tags match, the house is algebraizable. If they don't, it's not. It's a perfect "if and only if" match.

The Twist for 4-Dimensional Shapes

Now, imagine you try to build a house on a 4-dimensional shape (a hyper-donut, if you will). The authors say: "Hold your horses."

Here, the simple rule breaks down. Even if Tag 1 and Tag 2 are perfect and algebraic, the house might still refuse to be built with rigid rules.

The Obstacle: There is a hidden "ghost" in the machine. The authors proved that for 4-dimensional shapes, you also need to check two other things:
1. A number called the Pontryagin class (think of it as a measure of how much the fabric is "knotting" itself in a specific way).
2. A specific combination of the 4th tag and the first tag.
The Finding: Even if all your tags look algebraic, if these hidden knot-measures don't line up perfectly with a specific algebraic formula, the house cannot be built algebraically.
The Certainty: This is also proven. The authors didn't just guess; they constructed a specific mathematical "obstruction" (a barrier) that stops the conversion from happening. They even gave an example of a shape where this barrier is real and non-zero, proving that algebraic tags are not enough for 4D shapes.

The "Magic Circle" Example

To show how this works in the real world, the authors looked at a specific type of 4D shape: a 3D shape (like a sphere) multiplied by a circle (like a ring).

The Result: For this specific "Magic Circle" shape, they found that the hidden knot-measure (the Pontryagin class) must be zero for the house to be algebraizable.
The Takeaway: So, for these shapes, you need the algebraic tags plus the knot-measure to be exactly zero. If the knot-measure is anything else, the rigid blueprint doesn't exist.

Counting the Houses

Finally, the authors used these rules to count how many different "rigid houses" (algebraic vector bundles) can exist on these 4D shapes.

They broke the problem down into counting the algebraic tags and the knot-measures.
They found that the total number of these houses forms a specific group structure involving numbers like Z (integers) and Z/2 or Z/4 (groups of remainders).
The Certainty: They provided a proven formula (an isomorphism) that tells you exactly how to calculate the number of these houses based on the shape's properties.

What This Paper Does NOT Say

It does not say that all 4D shapes have this problem. It says there exists an obstruction. Some shapes might still work, but you can't assume they will just because the tags match.
It does not say that the "Magic Circle" example is the only place this happens. It's just a clear example where the math works out nicely.
It does not suggest that we can ignore the topological (rubber) blueprint. The algebraic blueprint must match the rubber one perfectly to exist.

The Bottom Line

For small shapes (up to 3D), checking the "tags" is enough to know if a flexible structure can be made rigid. For 4D shapes, checking the tags is necessary but not sufficient; you must also check the "knots" (Pontryagin classes). If the knots don't align with the algebraic rules, the rigid house simply cannot be built, no matter how perfect the tags look.

The authors have proven these rules using a powerful tool called "motivic homotopy theory," which is like a super-microscope that lets them see the deep connections between flexible shapes and rigid equations. They haven't just suggested this; they have built the mathematical bridge and walked across it to prove the destination.

Technical Summary: Algebraizability of Vector Bundles over Real Algebraic Varieties

1. Problem Statement

The paper investigates the algebraizability problem for real vector bundles. Let $X$ be a smooth affine real algebraic variety (in the sense of Bochnak, Coste, and Roy), and let $X(\mathbb{R})$ denote its set of real points equipped with the Euclidean topology. The central question is: given a topological real vector bundle $V$ over $X(\mathbb{R})$ , under what conditions is $V$ algebraizable? That is, when does there exist an algebraic vector bundle over $X$ (defined over a neighborhood of $X(\mathbb{R})$ ) whose topological realization is isomorphic to $V$ ?

While it is well-established that algebraic bundles possess "algebraic" characteristic classes (images of Chow groups under cycle class maps), the converse—whether algebraic characteristic classes suffice for algebraizability—has been a subject of long-standing study. Previous results confirmed the converse for low-dimensional varieties (dimension $\le 3$ ) and specific cases like spheres and projective spaces, often using topological or analytic methods. However, obstructions were known to exist in dimension 4. This paper aims to provide a complete characterization for smooth affine real algebraic varieties of dimensions 3 and 4 using motivic homotopy theory.

2. Methodology

The author employs motivic homotopy theory, specifically the framework developed by Morel, Asok, Fasel, and others. The core methodological pillars include:

Representability of Vector Bundles: Utilizing the theorem that for a smooth affine scheme $X$ , the set of algebraic vector bundles $\text{Vect}_n(X)$ is in bijection with the set of $\mathbb{A}^1$ -homotopy classes of maps $[X, BGL_n]_{\mathbb{A}^1}$ .
Postnikov Towers and Obstruction Theory: The problem of lifting a map to a classifying space is analyzed via relative Postnikov towers. The obstructions to algebraizability are identified as cohomology classes in the cohomology of $X$ with coefficients in the homotopy sheaves of the fiber of the map between classifying spaces.
Strongly $\mathbb{A}^1$ -Invariant Sheaves: The analysis relies heavily on the properties of sheaves that are strongly $\mathbb{A}^1$ -invariant (e.g., Milnor-Witt K-theory sheaves $\mathbf{K}^{MW}_*$ , Milnor K-theory sheaves $\mathbf{K}^M_*$ , and powers of the fundamental ideal $\mathbf{I}^n$ ).
Rost-Schmid Complexes: Explicit computations of cohomology groups are performed using the Rost-Schmid complex, which computes the cohomology of strongly $\mathbb{A}^1$ -invariant sheaves.
Real Cycle Class Maps: The connection between algebraic cohomology (Chow groups) and topological cohomology is established via real cycle class maps (signature maps), which relate algebraic characteristic classes to their topological counterparts (Stiefel-Whitney and Pontryagin classes).

3. Key Contributions and Results

3.1 Dimension 3: Characterization via Stiefel-Whitney Classes

For smooth affine real algebraic varieties of dimension 3, the paper establishes a complete necessary and sufficient condition for algebraizability.

Theorem 3.2.4: Let $X$ be a smooth affine real algebraic variety of dimension 3. A topological real vector bundle $V$ over $X(\mathbb{R})$ is algebraizable if and only if its first and second Stiefel-Whitney classes, $w_1(V)$ and $w_2(V)$ , are algebraic (i.e., lie in the image of the cycle class map from the Chow ring).
Methodology: The proof reduces the rank $\ge 3$ case to rank 3 by splitting off trivial bundles. It utilizes the Postnikov tower of $BGL_3$ , showing that obstructions vanish in dimension 3 because the relevant cohomology groups (involving $\pi^{\mathbb{A}^1}_i$ ) are zero or the cohomological dimension of $X$ prevents higher obstructions. For rank 2, the twisted Euler class is shown to be algebraic if $w_2$ is algebraic, leveraging properties of the cycle class map on twisted cohomology.

3.2 Dimension 4: New Obstructions

For dimension 4, the paper demonstrates that algebraic characteristic classes are not sufficient. New obstructions arise related to the first Pontryagin class $p_1$ and the fourth Stiefel-Whitney class $w_4$ .

Theorem 4.1.8: Let $X$ $X$ be a smooth affine real algebraic variety of dimension 4 with $X(\mathbb{R})$ $X (R)$ compact. Let $M_i$ $M_{i}$ be the orientable connected components and $N_j$ $N_{j}$ the non-orientable ones. A topological bundle $V$ $V$ is algebraizable if and only if there exist algebraic classes $c_1 \in CH^1(X)$ $c_{1} \in C H^{1} (X)$ , $c_2 \in CH^2(X)$ $c_{2} \in C H^{2} (X)$ , and $\alpha \in H^3_{\text{alg}}(X; \mathbb{Z}/2)$ $α \in H_{alg}^{3} (X; Z /2)$ such that:
1. $\gamma(c_1) = w_1(V)$ and $\gamma(c_2) = w_2(V)$ .
2. For each orientable component $M_i$ , the Pontryagin class satisfies $p_1(V|_{M_i}) = k_i(c_2 + c_1^2)$ , where $k_i: CH^2(X) \to H^4(M_i; \mathbb{Z})$ is a specific group homomorphism.
3. For each non-orientable component $N_j$ , the relation $(w_4(V) + w_1(V)w_3(V))|_{N_j} = \alpha|_{N_j} \cup w_1(N_j) + k'_j(c_2 + c_1^2)$ holds, where $k'_j$ is a specific map.
Significance: This theorem identifies that even if all Stiefel-Whitney classes are algebraic, the bundle may fail to be algebraizable if the Pontryagin class does not satisfy a specific algebraic constraint determined by $c_2$ and $c_1$ . The obstruction lies in the group $H^4(X, \pi^{\mathbb{A}^1}_3(F))$ , where $F$ is the fiber of the map $(c_2, c_4, p_1): BSL_\infty \to K(\mathbf{K}^M_2, 2) \times K(\mathbf{K}^M_4, 4) \times K(\mathbf{I}_4, 4)$ .

3.3 Applications

The paper applies these general theorems to specific geometric settings:

Product with a Circle (Theorem 4.2.2): Let $Y$ be a smooth affine real algebraic variety of dimension 3 with $Y(\mathbb{R})$ compact and orientable. Let $X = Y \times S^1$ . A topological bundle $V$ over $X(\mathbb{R})$ is algebraizable if and only if all Stiefel-Whitney classes are algebraic and the Pontryagin class vanishes ( $p_1(V) = 0$ ). This is derived from the fact that the relevant obstruction map $k$ vanishes because $CH^2(X)$ is 2-torsion while $H^4(X(\mathbb{R}); \mathbb{Z}) \cong \mathbb{Z}$ .
Structure of $\tilde{K}^0(X)$ (Theorem 4.2.5): For a dimension 4 variety $X$ with $X(\mathbb{R})$ compact and orientable, the paper provides an explicit decomposition of the reduced Grothendieck group of algebraic vector bundles, $\tilde{K}^0(X)$ . It establishes a short exact sequence relating $\tilde{K}^0(X)$ to a group $G$ constructed from algebraic cohomology and Chow groups. Specifically, if $X(\mathbb{R})$ is orientable, $\tilde{K}^0(X)$ is isomorphic to a direct sum of free and torsion parts determined by the dimensions of algebraic cohomology groups and the structure of $CH^2(X)$ .

4. Significance and Claims

The paper claims to resolve the algebraizability problem for smooth affine real algebraic varieties in dimensions 3 and 4 using the systematic machinery of motivic homotopy theory.

Unification: It unifies previous topological results (e.g., for $S^n$ or $\mathbb{R}P^n$ ) into a general framework based on $\mathbb{A}^1$ -homotopy classes and obstruction theory.
Precision in Dimension 4: It explicitly characterizes the failure of the "algebraic characteristic classes imply algebraizability" conjecture in dimension 4. The paper provides the precise algebraic formulas (involving $p_1$ and $w_4$ ) that must be satisfied, moving beyond the mere existence of counterexamples.
Computational Tools: By utilizing the Rost-Schmid complex and the properties of strongly $\mathbb{A}^1$ -invariant sheaves, the paper provides a computational method to determine the obstruction groups, which are shown to be non-trivial in dimension 4.

The author maintains a modest tone regarding the scope, noting that the definition of algebraic vector bundles is restricted to neighborhoods of $X(\mathbb{R})$ (due to the non-algebraically closed nature of $\mathbb{R}$ ) and that the specific maps $k'_j$ in the non-orientable case are not explicitly identified, though their existence and properties are established. The work serves as a foundational step in applying motivic homotopy theory to real algebraic geometry problems concerning vector bundles.