Table of Contents
Fetching ...

Étale cohomology of algebraic varieties over Stein compacta

Olivier Benoist

TL;DR

This work develops a robust bridge between étale cohomology of algebraic varieties over Stein algebras and singular cohomology of their analytifications, enabling a transfer of topological information to algebraic context. The authors establish absolute and relative Artin comparison theorems in the Stein setting, refine the comparison via the qfh topology, and prove cohomological-dimension bounds for fields of meromorphic functions near Stein compacta. They further exploit a Galois (G)–equivariant framework to derive quantitative real-analytic Hilbert 17th problem results, including sums of squares for nonnegative real-analytic functions on Stein spaces and Pfister-type bounds. These results illuminate the interplay between complex-analytic, real-analytic, and algebraic geometry, with concrete implications for cohomology, function fields, and real-analytic positivity problems.

Abstract

We prove a comparison theorem between the étale cohomology of algebraic varieties over Stein compacta and the singular cohomology of their analytifications. We deduce that the field of meromorphic functions in a neighborhood of a connected Stein compact subset of a normal complex space of dimension $n$ has cohomological dimension $n$. As an application of $\textrm{Gal}(\mathbb{C}/\mathbb{R})$-equivariant variants of these results, we obtain a quantitative version of Hilbert's 17th problem on compact subsets of real-analytic spaces.

Étale cohomology of algebraic varieties over Stein compacta

TL;DR

This work develops a robust bridge between étale cohomology of algebraic varieties over Stein algebras and singular cohomology of their analytifications, enabling a transfer of topological information to algebraic context. The authors establish absolute and relative Artin comparison theorems in the Stein setting, refine the comparison via the qfh topology, and prove cohomological-dimension bounds for fields of meromorphic functions near Stein compacta. They further exploit a Galois (G)–equivariant framework to derive quantitative real-analytic Hilbert 17th problem results, including sums of squares for nonnegative real-analytic functions on Stein spaces and Pfister-type bounds. These results illuminate the interplay between complex-analytic, real-analytic, and algebraic geometry, with concrete implications for cohomology, function fields, and real-analytic positivity problems.

Abstract

We prove a comparison theorem between the étale cohomology of algebraic varieties over Stein compacta and the singular cohomology of their analytifications. We deduce that the field of meromorphic functions in a neighborhood of a connected Stein compact subset of a normal complex space of dimension has cohomological dimension . As an application of -equivariant variants of these results, we obtain a quantitative version of Hilbert's 17th problem on compact subsets of real-analytic spaces.
Paper Structure (35 sections, 51 theorems, 44 equations)

This paper contains 35 sections, 51 theorems, 44 equations.

Key Result

Theorem 1

Let $M$ be a normal real-analytic variety of pure dimension $n$. Let $K\subset M$ be compact. Let $f:M\to \mathbb R$ be a nonnegative real-analytic function. Then there exists an open neighborhood $U$ of $K$ in $M$ such that $f|_U$ is a sum of $2^n$ squares of real-analytic meromorphic functions on

Theorems & Definitions (96)

  • Theorem 1: Theorem \ref{['sosreal']}
  • Corollary 2
  • Theorem 3: Theorem \ref{['soscx']}
  • Theorem 4: Corollary \ref{['cdfieldcor']} and Remark \ref{['remcd']} (ii)
  • Theorem 5: Theorem \ref{['Artincompcx']}
  • Theorem 6: Theorem \ref{['relArtin']}
  • Lemma 1.1
  • proof
  • Proposition 1.2
  • proof
  • ...and 86 more