Cohomological Chow Groups of codimension one of varieties with isolated singularities

This paper computes cohomological Chow groups of codimension one for varieties with isolated singularities, establishing results for higher-dimensional varieties with contractible dual complexes and for 3-dimensional varieties under the weaker condition that the second cohomology of the dual complex vanishes.

The Gist

Here is an explanation of the paper "Cohomological Chow Groups of Codimension One of Varieties with Isolated Singularities" by Diosel López-Cruz, translated into simple, everyday language with creative analogies.

The Big Picture: Smoothing Out the Cracks

Imagine you are an architect looking at a beautiful, complex building (a mathematical variety). Most of the building is perfectly smooth and well-constructed. However, there are a few spots where the walls have collapsed or the corners are jagged. These are isolated singularities—tiny, specific points where the geometry breaks down.

Mathematicians love to measure things about buildings: how many rooms they have, how they are connected, and how they twist. In algebraic geometry, they use tools called Chow Groups to count and classify these features.

• The Problem: Standard counting tools work great on smooth buildings. But when you try to count features on a building with a jagged, broken corner, the tools break. They give you nonsense numbers or get stuck.
• The Solution: This paper introduces a special "repair kit" called Cohomological Chow Groups. It's a way to count the features of a broken building by pretending it's been fixed, but keeping a detailed record of exactly how it was broken.

The Metaphor: The "Magic Blueprint" (Hyperresolution)

How do you study a broken building without getting hurt by the jagged edges? You don't look at the broken building directly. Instead, you create a Magic Blueprint (mathematically called a semi-simplicial hyperresolution).

Imagine you have a crumpled piece of paper (the singular variety). You can't read the text on it. So, you unfold it, tape it flat, and then take a photo of every single layer of the tape and the paper underneath.

1. The Smooth Layers: You replace the broken spot with a perfect, smooth patch (the resolution).
2. The Boundary: Where the patch meets the original paper, there is a seam. This seam is a Normal Crossing Divisor (a fancy way of saying a collection of smooth lines or surfaces that cross each other neatly).
3. The Map: You then map out how all these seams connect to each other. If you have three seams meeting at a point, you draw a triangle. If two meet, you draw a line. This map is called the Dual Complex (or $\Gamma$).

The paper's main job is to figure out how to count the "rooms" (codimension one features) of the original broken building by looking at:

1. The smooth patch.
2. The map of the seams (the Dual Complex).

The Main Discoveries

The author proves three main things, which we can think of as rules for how the "brokenness" affects the count.

1. The "Tree" Rule (For 3D Buildings)

If you have a 3-dimensional building with a broken corner, the author looks at the map of the seams (the Dual Complex).

• The Condition: If this map looks like a tree (a shape with no loops, where you can get from any point to any other without circling back), then the math is very simple.
• The Result: The "brokenness" doesn't create any weird, extra hidden rooms. The count of features is almost exactly the same as if the building were perfect, except for a few specific adjustments at the very bottom of the count (degrees -2, -1, 0, 1).
• Analogy: If the seams of your patch form a simple tree, the building's "soul" (its cohomological group) is very stable. You don't have to worry about complex loops confusing the count.

2. The "Contractible" Rule (For Taller Buildings)

What if the building is 4, 5, or even 100 dimensions tall?

• The Condition: The author asks: What if the map of the seams is contractible? In math terms, this means the map can be shrunk down to a single point without tearing. Think of a rubber ball; you can shrink it to a dot.
• The Result: If the seam-map is "shrinkable," the author can give you a precise formula for every level of the count. It turns out the count depends entirely on the smooth patch and the "shape" of the seams.
• Analogy: If the blueprint of the seams is simple enough to be squashed into a dot, the broken building behaves very predictably. You can calculate exactly how many "rooms" exist at every level of complexity.

3. The "Exact Sequence" (The Accounting Equation)

The paper provides a specific equation (an exact sequence) that acts like a balance sheet.

• The Equation:
  Total Broken Building = (Smooth Patch) - (Seam Map) + (Adjustments)
• What it means: To find the true count of the broken building, you take the count of the smooth patch, subtract the count of the seams, and add back a tiny correction factor. This allows mathematicians to calculate the properties of a messy, broken object by using the clean properties of the smooth parts.

Why Does This Matter?

In the real world, things are rarely perfect. Crystals have defects, proteins fold in messy ways, and data sets have noise. In mathematics, "singularities" (broken points) are everywhere.

This paper is like a new calculator for broken things.

• Before this, if you had a 3D shape with a broken point, you might not know how to count its features if the "seams" were complicated.
• Now, if you check the "shape of the seams" (the Dual Complex) and see that it's a simple tree or a shrinkable blob, you can instantly know the answer.

Summary in One Sentence

This paper teaches us how to count the hidden features of a broken mathematical shape by replacing the break with a smooth patch, mapping the seams where they connect, and proving that if those seams form a simple shape (like a tree), the math becomes surprisingly easy and predictable.

Technical Summary

Here is a detailed technical summary of the paper "Cohomological Chow Groups of Codimension One of Varieties with Isolated Singularities" by Diosel López-Cruz.

1. Problem Statement

The paper addresses the computation of cohomological Chow groups (also known as motivic cohomology) for singular algebraic varieties, specifically focusing on codimension one ($r=1$).

• Context: For smooth varieties, higher Chow groups $CH_r(X, m)$ are well-defined via Bloch's cycle complexes and relate to motivic cohomology. However, for singular varieties, higher Chow groups behave as a Borel-Moore homology theory rather than a cohomology theory.
• The Gap: A cohomological version for singular varieties was constructed by Hanamura using cubical hyperresolutions (semi-simplicial schemes of smooth varieties). While the theory is established, explicit computations for specific classes of singular varieties remain challenging.
• Specific Focus: The author investigates complex projective varieties $X$ of dimension $d \geq 3$ with isolated singularities ($X_{sing}$ is a finite set of points). The goal is to determine the structure of the groups $CHC^1(X, m)$ (cohomological Chow groups of codimension 1) and identify the conditions under which they vanish or are non-zero.

2. Methodology

The paper employs a recursive geometric approach based on resolution of singularities and spectral sequences.

A. Semi-Simplicial Hyperresolutions

The core tool is the construction of a semi-simplicial hyperresolution $a: X_\bullet \to X$.

• Construction: Starting with a resolution of singularities $p: \tilde{X} \to X$ where the exceptional fiber $E = p^{-1}(X_{sing})$ is a Simple Normal Crossing (SNC) divisor, the author recursively resolves the singularities of $E$ and the singular locus $X_{sing}$.
• Structure: This yields a semi-simplicial scheme $X_\bullet = \{X_0, X_1, \dots, X_n\}$ where each $X_p$ is a smooth variety. For isolated singularities, the length of this resolution is bounded by the dimension of $X$.
• Cohomological Cycle Complex: The cohomological Chow groups are defined as the homology of the total complex of a double complex formed by Bloch's cycle complexes $Z_r(X_p, \bullet)$ of the smooth components, with horizontal differentials induced by face maps.

B. Spectral Sequences

The author utilizes a fourth-quadrant spectral sequence associated with the hyperresolution:
$$E_1^{p,q}(r) := CH_r(X_p, -q) \implies CHC^r(X, -p-q)$$
This allows the computation of the singular variety's groups by analyzing the smooth components $X_p$ and the differentials between them.

C. Dual Complexes

A crucial geometric invariant introduced is the dual complex $\Gamma(E)$ associated with the SNC divisor $E$.

• Definition: Vertices correspond to irreducible components of $E$, edges to intersections of two components, and higher-dimensional cells to higher-order intersections.
• Key Insight: The paper establishes a link between the cohomological Chow groups of the exceptional divisor $E$ and the cohomology of the dual complex $\Gamma(E)$. Specifically, for codimension 1, the complex of higher Chow groups of the components of $E$ is identified with the chain complex of $\Gamma(E)$ tensored with $\mathbb{C}^*$.

3. Key Contributions and Results

The paper provides explicit computations for $CHC^1(X, m)$ under specific topological conditions on the dual complex $\Gamma(E)$.

A. General Properties for Isolated Singularities

• Vanishing in Positive Degrees: For any variety $X$, $CHC^1(X, m) = 0$ for $m > 1$.
• Vanishing in Negative Degrees: Unlike smooth varieties, these groups can be non-zero in negative degrees. The paper determines the precise range of non-vanishing degrees based on the dimension $d$.

B. The Case of Threefolds ($d=3$)

Theorem 3.8: Let $X$ be a 3-dimensional irreducible projective variety with isolated singularities. If the dual complex $\Gamma(E)$ satisfies $H_2(\Gamma(E)) = 0$, then:

• $CHC^1(X, m) = 0$ for $m \notin \{-2, -1, 0, 1\}$.
• $CHC^1(X, 1) \cong \mathbb{C}^*$.
• There is a short exact sequence:
  $$0 \to CHC^1(X) \to CH_1(\tilde{X}) \to CHC^1(E) \to CHC^1(X, -1) \to 0$$
• Significance: The condition $H_2(\Gamma)=0$ is weaker than contractibility but sufficient to collapse the spectral sequence, isolating the contribution of the dual complex's homology to the Chow groups.

C. The Case of Higher Dimensions ($d \geq 3$)

Theorem 3.10: Let $X$ be a $d$-dimensional complex projective variety with isolated singularities. If the dual complex $\Gamma(E)$ is contractible, then:

• $CHC^1(X, m) \neq 0$ for $m \in \{1, 0, -1, \dots, d-1\}$.
• $CHC^1(X, 1) \cong \mathbb{C}^*$.
• There is a short exact sequence:
  $$0 \to CHC^1(X) \to CH_1(\tilde{X}) \to CHC^1(E) \to CHC^1(X, -1) \to 0$$
• Proposition 3.9: For the exceptional divisor $E$ itself, if $\Gamma(E)$ is contractible, the groups $CHC^1(E, m)$ are isomorphic to the homology of the complex of higher Chow groups of the components of $E$ (shifted by degree).

4. Significance and Implications

1. Explicit Computations: The paper moves beyond abstract existence theorems to provide concrete descriptions of motivic cohomology groups for singular varieties, a notoriously difficult area in algebraic geometry.
2. Topological Connection: It rigorously demonstrates that the cohomological Chow groups of a singular variety with isolated singularities are deeply governed by the topology of the dual complex of the exceptional divisor in a resolution. This bridges algebraic cycle theory with combinatorial topology.
3. Refinement of Conditions: The work distinguishes between the conditions required for different dimensions. For 3-folds, the vanishing of the second homology group ($H_2=0$) is sufficient, whereas for higher dimensions, the contractibility of the dual complex is used to ensure the spectral sequence collapses cleanly.
4. Negative Degree Behavior: The results clarify the behavior of these groups in negative degrees, showing that for singular varieties, non-trivial information persists in degrees $m < 0$, unlike the smooth case where they vanish.

In summary, López-Cruz successfully computes the codimension-one cohomological Chow groups for varieties with isolated singularities by reducing the problem to the study of the dual complex of the exceptional divisor, providing exact sequences that relate the singular variety's invariants to those of its smooth resolution.