Gabber's presentation lemma over noetherian domains

Neeraj Deshmukh, Amit Hogadi, Girish Kulkarni, Suraj Yadav

TL;DR

This paper generalizes Gabber's presentation lemma, a fundamental tool in algebraic geometry, from Dedekind domains to general noetherian domains with infinite residue fields. This generalization is crucial for extending results in A¹-homotopy theory, specifically A¹-connectivity properties, to a broader class of base schemes.

What Problem Does This Paper Address?

Gabber's presentation lemma, originally formulated for infinite fields, provides a way to "straighten out" certain geometric situations, particularly by showing that a smooth scheme can be locally presented as a graph of a finite map and an étale map. This lemma is a cornerstone for understanding A¹-homotopy theory, a field that uses a different notion of "space" (sheaves of spectra on schemes) and "homotopy" (related to the affine line A¹).

Previous work by Schmidt and Strunk generalized this lemma to Dedekind domains with infinite residue fields. However, a significant gap remained: how to extend this powerful tool to more general noetherian domains, which are ubiquitous in algebraic geometry. This paper aims to bridge that gap, enabling the application of A¹-homotopy theory to a wider range of base spaces.

Methods

The authors build upon the existing proofs for Dedekind domains and adapt them to the more general setting of noetherian domains. The core of their approach involves:

  • Fiber-wise denseness: A key technical result (Theorem 2.1) is established, which is a weaker version of a "projective closure" lemma from Kai. This lemma ensures that a scheme can be embedded into a projective space such that it remains "dense" in certain components of its fibers. This is crucial for constructing the required maps.
  • Induction and localizations: The proofs often rely on induction on the dimension of the base scheme and on taking Nisnevich neighborhoods. Nisnevich neighborhoods are a type of open set that allows for localizing the problem without losing essential geometric information.
  • Decomposition into finite and étale maps: The central idea of Gabber's lemma is to decompose a local geometric situation into a finite map (which "collapses" parts of the space) and an étale map (which locally looks like a projection). The paper carefully constructs these maps by leveraging techniques like Stein factorization and by carefully choosing hyperplanes in projective spaces.
  • Radicial maps: The lemma also requires constructing "radicial" maps, which are a type of map that is both injective and has trivial residue field extensions.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Generalization is possible: The most significant insight is that the core arguments of Gabber's presentation lemma can be adapted to a broader class of base schemes (noetherian domains with infinite residue fields) than previously thought. This is not a trivial extension and requires careful technical work.
  • Weaker denseness is sufficient: The paper demonstrates that a slightly weaker condition of "fiber-wise denseness" in projective closures is enough to prove the generalized lemma, which is a valuable technical simplification.
  • Enabling broader A¹-homotopy theory: The primary impact of this generalization is its direct application to proving A¹-connectivity results for a wider range of base schemes. This means that sophisticated tools from A¹-homotopy theory can now be applied to more general algebraic varieties.

Key Results

The paper's main contributions are:

  • Theorem 1.1 (Gabber's Presentation Lemma): This theorem provides the generalized version of Gabber's lemma for smooth, irreducible, equi-dimensional S-schemes over a noetherian domain S with infinite residue fields. It guarantees the existence of a local presentation of a closed subscheme Z within X as a combination of finite and étale maps.
  • Theorem 1.2 (Stable A¹-Connectivity): As a direct consequence of Theorem 1.1, the paper proves that any noetherian domain S of dimension d with infinite residue fields has the shifted stable A¹-connectivity property. This is a significant result in A¹-homotopy theory, indicating how "connected" certain spectral constructions are.

Limitations and Future Work

  • Infinite residue fields: The current generalization still requires the residue fields of the base scheme to be infinite. The authors mention that they are "currently working on removing the condition of residue fields being infinite," suggesting this is a key direction for future research.
  • Dedekind vs. General Domains: While the paper generalizes to general noetherian domains, the proofs often build upon results for Dedekind domains. Further research might explore more direct approaches for the general case.
  • Applications: The paper demonstrates one application to A¹-connectivity. Further exploration of how this generalized lemma can be used to prove other results in A¹-homotopy theory and related areas is a natural next step.

Summary updated

Abstract

Following Schmidt and Strunk, we give a proof of Gabber presentation lemma over a noetherian domain with infinite residue fields.
BibTeX
@article{deshmukh2020gabber,
  author = {Neeraj Deshmukh and Amit Hogadi and Girish Kulkarni and Suraj Yadav},
  title = {{Gabber's presentation lemma over noetherian domains}},
  journal = {Journal of Algebra},
  year = {2020},
  doi = {10.1016/j.jalgebra.2020.10.036},
}
Figure