🔢 mathematics

On the bisections of a local Lie grpoupod

This paper investigates the local Lie group structure formed by the admissible bisections of a local Lie groupoid over a compact manifold, explores its relationship with the Lie algebra of the associated Lie algebroid, and demonstrates that the globalizability of the groupoid implies the globalizability of its bisection group.

Original authors: Navya K Nair, P G Romeo

Published 2026-06-18

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Navya K Nair, P G Romeo

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 trying to build a giant, complex machine (like a spaceship) from a set of blueprints. In the world of advanced mathematics, specifically geometry, these blueprints are called Lie groupoids. They describe how different parts of a shape can move, rotate, or transform into one another.

However, sometimes the blueprints are incomplete. You might only have instructions for how the machine works in a small, safe room right next to the starting point. You know how to take a few steps, but you don't know if you can keep going forever without the machine falling apart or the instructions contradicting themselves. This is called a local Lie groupoid. It's a "local" version of the full machine.

The big question mathematicians ask is: Can we extend these local instructions to build the whole global machine? This is called the "globalization" problem.

The Problem: The "Local" vs. "Global" Gap

In the past, mathematicians knew that for simple, single-point machines (called Lie groups), there was a rule: if the instructions are perfectly consistent no matter how many times you combine them (a property called "global associativity"), then you can build the whole machine.

But for complex, multi-part machines (Lie groupoids), this was harder to prove. The paper by Nair and Romeo tackles this by looking at a specific tool: Bisections.

The Tool: "Bisections" as the Control Panel

Think of a bisection as a "control panel" or a "snapshot" of the machine.

If the machine is a fleet of drones flying over a city, a bisection is a specific command that tells every drone exactly where to go, ensuring that for every starting point in the city, there is exactly one drone landing there.
The authors look at the collection of all possible valid control panels (bisections) for a local machine.

They discovered something amazing: The collection of all these control panels forms its own smaller machine (a local Lie group).

The Main Discovery: The "Shadow" Machine

The paper proves three main things using this "control panel" idea:

The Control Panel is a Machine: If you have a local Lie groupoid (a partial machine), the space of all its valid control panels (bisections) naturally forms a "local Lie group." It has its own rules for combining commands and undoing them, just like a machine does.
The Connection: There is a direct, mathematical link between the "engine" of this new control-panel machine and the "engine" of the original partial machine. They are two sides of the same coin.
The Big Reveal (Globalization): This is the most important part. The authors prove that if the original partial machine (the local Lie groupoid) can be extended into a full, global machine, then its control-panel machine (the local Lie group of bisections) can also be extended into a full machine.
- Analogy: Imagine you have a puzzle with a few missing pieces. If you can prove the puzzle can be completed, then the "instruction manual" you wrote for the puzzle pieces can also be completed. The ability to finish the big machine guarantees the ability to finish the control panel machine.

The "Consistency" Rule

The paper relies on a concept called associativity.

Imagine you are stacking blocks. If you stack A, then B, then C, does it matter if you group them as (A+B)+C or A+(B+C)?
In a "local" machine, you might only be sure this works for small stacks.
The paper shows that if the original machine is consistent no matter how high you stack the blocks (globally associative), then the machine made of control panels is also consistent.
Conversely, if the control panels are perfectly consistent, and every part of the original machine is covered by at least one control panel, then the original machine is also consistent and can be built globally.

A Concrete Example: The Sphere

The authors use a specific example involving a sphere (like the Earth).

They create a machine where points on the sphere are connected by paths, and the "height" of the path matters.
They show that even though the rules for combining these paths are only defined locally (you can't always go all the way around the sphere without hitting a "no-entry" zone), the collection of all valid ways to map the sphere to itself (the bisections) forms a smooth, well-behaved group.
They prove that because the rules for the sphere machine are consistent, the rules for the mapping machine are also consistent, allowing both to be "globalized."

Summary

In simple terms, this paper builds a bridge between two worlds:

The world of partial geometric machines (local Lie groupoids).
The world of collections of control panels (local Lie groups of bisections).

The authors show that these two worlds are tightly linked. If one can be expanded into a complete, global structure, the other can too. They provide a new way to check if a complex geometric structure can be "finished" by looking at the consistency of its control panels.

Technical Summary: On the Bisections of a Local Lie Groupoid

Problem Statement
The paper addresses the integration theory of Lie algebroids. While every Lie algebroid admits a local integration to a local Lie groupoid (as shown by Cabrera, Mărcuţ, and Salazar), the existence of a global Lie groupoid integration is obstructed by global monodromy groups (Crainic and Fernandes). Consequently, local Lie groupoids serve as the natural infinitesimal-to-global bridge. A central problem in this theory is "globalizability": determining when a local Lie groupoid can be extended to a global Lie groupoid.

Parallel to this, the theory of bisections provides a link between Lie groupoids and infinite-dimensional Lie theory. Schmeding and Wockel established that the space of bisections of a global Lie groupoid carries a natural infinite-dimensional Lie group structure. This paper investigates the analogous structure for local Lie groupoids, specifically asking:

Does the space of admissible bisections of a local Lie groupoid form a local Lie group?
What is the relationship between the Lie algebra of this local Lie group and the sections of the associated Lie algebroid?
How does the globalizability of a local Lie groupoid relate to the globalizability of its associated local Lie group of bisections?

Methodology and Framework
The authors work within the framework of infinite-dimensional manifold theory, specifically utilizing locally convex, locally metrizable topological vector spaces. The study focuses on local Lie groupoids $G$ over a compact manifold $M$ that admit an adapted local addition. This technical condition is crucial for constructing manifold structures on spaces of smooth maps ( $C^\infty(M, G)$ ) and sections.

The methodology proceeds through the following steps:

Construction of the Bisection Space: The authors define a bisection of a local Lie groupoid as a smooth section $\sigma: M \to G$ of the source map $\alpha$ such that $\beta \circ \sigma$ is a diffeomorphism. They construct the manifold structure on the space of such bisections, denoted $\text{Bis}(G)$ , by identifying it as an open submanifold of the space of sections $\Gamma(M \leftarrow_\alpha G)$ .
Group Structure Definition: Multiplication and inversion are defined pointwise using the local multiplication and inversion of $G$ , adjusted by the diffeomorphism property of the bisections.
Lie Algebra Correspondence: The tangent space at the identity of $\text{Bis}(G)$ is analyzed to establish an isomorphism with the space of sections of the Lie algebroid $L(G)$ .
Globalization Analysis: The paper employs Mal'cev's characterization of globalizability for local Lie groups (global associativity) and the analogous result for local Lie groupoids (established by Fernandes, Michiels, and Zhang) to derive conditions under which the globalizability of one structure implies the other.

Key Contributions and Results

Local Lie Group Structure on Bisections (Theorem 1.3 / 3.3):
The paper proves that for a locally convex, locally metrizable local Lie groupoid $G$ over a compact manifold $M$ admitting an adapted local addition, the set of all admissible bisections $\text{Bis}(G)$ forms a local Lie group. The proof relies on the smoothness of the evaluation map and the fact that the space of diffeomorphisms is an open submanifold of the space of smooth maps.
Lie Algebra Isomorphism (Theorem 4.3):
The authors establish that the Lie algebra of the local Lie group $\text{Bis}(G)$ is anti-isomorphic to the space of sections $\Gamma(L(G))$ of the Lie algebroid associated with $G$ . This is achieved by constructing a map $\phi_G$ from the tangent space at the identity of $\text{Bis}(G)$ to the sections of $L(G)$ and proving that it preserves the Lie bracket structure via $\gamma$ -relatedness of vector fields.
Globalization Implications (Theorem 5.2, Corollary 5.3, Theorem 5.5, Theorem 5.7):
- Forward Implication: If a local Lie groupoid $G$ is globally associative (and thus globalizable, assuming strong connectivity), then its associated local Lie group of bisections $\text{Bis}(G)$ is also globally associative and, if connected, globalizable.
- Reverse Implication: Under the condition that every arrow in $G$ is contained in a global bisection, if $\text{Bis}(G)$ is globally associative, then $G$ is globally associative.
- Embedding Result: If $G$ is globalizable with a globalization $\tilde{G}$ , then $\text{Bis}(G)$ embeds as an open local Lie subgroup of the global Lie group $\text{Bis}(\tilde{G})$ . Consequently, the globalizability of $G$ implies the globalizability of $\text{Bis}(G)$ .

Significance and Claims
The paper claims to provide a rigorous infinite-dimensional Lie theoretic framework for local Lie groupoids, mirroring the well-established theory for global Lie groupoids. By constructing the local Lie group structure on bisections and linking its Lie algebra to the underlying Lie algebroid, the work extends the "bridge" between infinitesimal and global structures to the local setting.

The significance lies in the characterization of globalizability: the paper demonstrates that the obstruction to globalizing a local Lie groupoid is intimately tied to the global associativity of its bisection group. Specifically, it establishes that for strongly connected local Lie groupoids where every arrow lies in a global bisection, the globalizability of the groupoid and the globalizability of its bisection group are equivalent conditions. This result unifies the integration problems of Lie algebroids and the structural theory of infinite-dimensional Lie groups of bisections within the local context.