An adjunction inequality for Real embedded surfaces

This paper establishes necessary and sufficient cohomological conditions for representing classes by Real embedded surfaces in 4-manifolds and derives two versions of an adjunction inequality for their genus when Real Seiberg-Witten invariants are non-zero, demonstrating that the minimal genus for Real surfaces can exceed that of arbitrary embedded surfaces.

The Gist

Imagine you have a giant, four-dimensional piece of clay. In mathematics, this is called a 4-manifold. Now, imagine you have a magical mirror that can fold this clay in half perfectly. This folding action is called a Real structure (denoted by $\sigma$). When you fold the clay, some parts of it might land exactly on top of themselves, while other parts get flipped over.

This paper, written by David Baraglia, is about drawing surfaces (like sheets of paper or soap bubbles) inside this 4D clay, but with a special rule: the surface must respect the fold. If you fold the clay, the surface must flip over onto itself. Mathematicians call these Real embedded surfaces.

Here is the breakdown of the paper's big ideas, using simple analogies:

1. The "Can I Draw It?" Problem (The Existence Question)

Before you can draw a picture, you need to know if the paper is big enough and if the rules allow it.

• The Question: If I want to draw a surface that represents a specific shape or "class" in this 4D clay, can I actually draw a Real surface (one that respects the fold)?
• The Discovery: The author proves a strict rule. You can only draw such a surface if your shape has a special "symmetry signature." In math terms, the shape must change sign when you look at it through the mirror. If it doesn't have this signature, no amount of folding will let you draw it.
• The Analogy: Imagine trying to fold a piece of paper so that a drawing of a cat lands on top of itself, but the cat's head is now its tail. If the drawing isn't symmetric in a specific way, it's impossible. This paper gives the exact checklist to see if your drawing is possible.

2. The "How Big Does It Have to Be?" Problem (The Genus Question)

Once you know you can draw the surface, the next question is: How complicated does it have to be?

• The Concept: In math, the "complexity" of a surface is measured by its genus (the number of holes). A sphere has 0 holes (genus 0). A donut has 1 hole (genus 1). A pretzel has 3 holes (genus 3).
• The Goal: What is the minimum number of holes a Real surface needs to have to fit in the clay?
• The Old Way: For normal surfaces (ignoring the fold), mathematicians have a famous rule called the Adjunction Inequality. It says: "If the surface is big enough and the 4D clay is 'twisty' enough, the surface must have a certain number of holes. You can't cheat and make it a simple sphere."

3. The New Rule: The "Real" Adjunction Inequality

Baraglia proves a new version of this rule specifically for surfaces that respect the fold.

• The Twist: He uses a powerful tool called Real Seiberg–Witten invariants. Think of these as a "magic detector" that can sense the hidden twists and turns of the 4D clay. If the detector goes off (the invariant is non-zero), it forces any Real surface to be complex.
• The Result:
  • If the clay is "twisty" enough (mathematically, if $b_+ > 1$), and the detector goes off, then you cannot draw a simple sphere (genus 0) or a simple donut (genus 1) for certain shapes.
  • The surface must have a minimum number of holes, calculated by a specific formula involving how the surface intersects itself.

4. The Big Surprise: Real Surfaces are "Harder" to Draw

This is the most exciting part of the paper.

• The Scenario: Imagine you have a 4D shape made by gluing two other shapes together (a "connected sum").
  • Normal Surfaces: If you try to draw a normal surface here, the "magic detector" usually says "0" (it vanishes). This means the old rules don't apply, and you might be able to draw a very simple surface (like a sphere).
  • Real Surfaces: But, if you look at the Real version of the detector, it might say "Non-zero!"
• The Consequence: This creates a situation where:
  • You can draw a simple sphere (0 holes) if you ignore the fold.
  • But if you must respect the fold (Real surface), the new rules force you to draw a complex pretzel (many holes).
• The Analogy: Imagine you are trying to tie a knot in a rope.
  • If you can move the rope freely in 3D space, you can untie it easily (it's a simple loop).
  • But if you are forced to tie the knot while wearing thick gloves that restrict your movement (the "Real" constraint), you might find it impossible to make a simple loop. You are forced to make a complicated, knotty mess.
  • Baraglia's paper proves that for certain 4D shapes, the "Real" constraint forces the surface to be much more knotty (higher genus) than the "unconstrained" version.

5. Why Does This Matter?

• Real Algebraic Geometry: This connects to real-world shapes found in nature and art (like real algebraic curves). It helps mathematicians understand the limits of what shapes can exist in the real world versus the complex world.
• Topology: It shows that symmetry (the fold) can fundamentally change the geometry of a space. Sometimes, adding a symmetry constraint makes things harder to do, forcing surfaces to be more complex than they would be otherwise.

Summary in One Sentence

David Baraglia discovered that if you try to draw a shape in a 4D universe that respects a mirror fold, the shape is often forced to be much more complex (have more holes) than it would be if you didn't have to respect the fold, and he provided the exact mathematical formula to calculate just how complex it must be.

Technical Summary

Here is a detailed technical summary of the paper "An Adjunction Inequality for Real Embedded Surfaces" by David Baraglia.

1. Problem Statement

The paper addresses two fundamental problems in the topology of 4-manifolds equipped with a Real structure (an orientation-preserving smooth involution $\sigma$):

1. Existence: Given a cohomology class $\alpha \in H^2(X; \mathbb{Z})$, under what conditions can it be represented by a Real embedded surface (an embedded surface $\Sigma$ such that $\sigma(\Sigma) = \Sigma$ and $\sigma$ reverses the orientation of $\Sigma$)?
2. Minimal Genus: If such a surface exists, what is the minimum genus $g_{\min}^R(\alpha)$ among all Real embedded surfaces representing $\alpha$?

The author aims to establish an adjunction inequality for these Real surfaces, analogous to the classical adjunction inequality derived from Seiberg–Witten theory for ordinary embedded surfaces. A key motivation is to demonstrate that the minimal genus for Real surfaces can be strictly larger than the minimal genus for arbitrary (non-Real) embedded surfaces, even when the ordinary Seiberg–Witten invariants vanish.

2. Methodology

The paper employs a combination of equivariant topology, gauge theory, and geometric constructions:

• Equivariant Cohomology: The author utilizes the equivariant cohomology group $H^2_{\mathbb{Z}_2}(X; \mathbb{Z}^-)$, where $\mathbb{Z}^-$ denotes the local system on which $\sigma$ acts as $-1$. This is used to characterize the existence of Real surfaces via the "forgetful map" to ordinary cohomology.
• Real Line Bundles and Sections: The paper establishes a correspondence between Real embedded surfaces and the zero loci of Real sections of Real line bundles. A Real line bundle is a complex line bundle equipped with an anti-linear involution lifting $\sigma$.
• Real Seiberg–Witten Theory: The core analytical tool is the Real Seiberg–Witten invariant ($SW_R$). The author reviews the definition of these invariants (both mod 2 and integer versions) for manifolds with involutions, focusing on the behavior of the moduli space of solutions to the Seiberg–Witten equations restricted to Real configurations.
• Neck-Stretching and Blow-up Techniques:
  • Neck-stretching: Used to analyze solutions near the surface $\Sigma$ to derive the inequality for non-negative self-intersection.
  • Equivariant Blow-ups: The author develops specific blow-up formulas for Real structures. This includes blowing up at fixed points (equivariant connected sum with $\mathbb{C}P^2$) and at pairs of non-fixed points (connected sum with two copies of $\mathbb{C}P^2$). These operations allow the reduction of surfaces with positive self-intersection to those with zero self-intersection while preserving the non-vanishing of the Real Seiberg–Witten invariant.
• Connected Sum Formulas: The paper utilizes connected sum formulas for Real Seiberg–Witten invariants to construct examples where ordinary invariants vanish but Real invariants do not.

3. Key Contributions and Results

A. Existence of Real Embedded Surfaces

Theorem 1.1 & 2.3: A class $\alpha \in H^2(X; \mathbb{Z})$ can be represented by a Real embedded surface if and only if $\alpha$ lies in the image of the forgetful map:
$$H^2_{\mathbb{Z}_2}(X; \mathbb{Z}^-) \to H^2(X; \mathbb{Z})$$
Furthermore, if $b_1(X) = 0$ and $\sigma$ is not free, this image is exactly the set of classes satisfying $\sigma^*(\alpha) = -\alpha$.

B. Adjunction Inequalities

The paper proves two versions of the adjunction inequality, assuming $b_+(X)^{-\sigma} > 1$ (dimension of the $-1$ eigenspace of $\sigma$ on $H^+(X)$) and that the Real Seiberg–Witten invariant is non-zero.

1. Non-negative Self-Intersection (Theorem 1.3):
   For a Real surface $\Sigma$ of genus $g > 0$ with $[\Sigma]^2 \geq 0$:
   $$2g - 2 \geq |\langle c(s), [\Sigma] \rangle| + [\Sigma]^2$$
   If $g=0$, then $[\Sigma]^2 \leq 0$. If $\sigma$ is not free and $[\Sigma]$ is non-torsion, then $[\Sigma]^2 < 0$.

2. Arbitrary Self-Intersection (Theorem 1.4):
   For a Real surface $\Sigma$ of genus $g$ where $\sigma$ does not act freely on $\Sigma$:
   $$2g \geq |\langle c(s), [\Sigma] \rangle| + [\Sigma]^2$$
   Note: This version requires the integer invariant $SW_{R,\mathbb{Z}}$ to be defined and non-zero.

C. Separation of Real and Ordinary Minimal Genus

The most significant application of these results is the construction of 4-manifolds where the minimal genus for Real surfaces is strictly higher than for ordinary surfaces.

Theorem 1.5 & 6.7: The author constructs examples (specifically connected sums of $\mathbb{C}P^2$, $\overline{\mathbb{C}P^2}$, $S^2 \times S^2$, and $K3$ surfaces) admitting an involution $\sigma$ such that:

• The ordinary Seiberg–Witten invariants vanish (so the classical adjunction inequality provides no lower bound).
• The Real Seiberg–Witten invariants are non-zero.
• For a class $u$ with $u^2 \geq 0$, there exists a smooth embedded surface of genus $\leq u^2/2$.
• However, any Real embedded surface representing $u$ must have genus $\geq u^2/2 + 1$.

This proves that the topological constraints imposed by the Real structure can force the genus to be strictly larger than the unconstrained minimum.

4. Significance

• New Topological Invariants: The paper provides a robust framework for studying surfaces in 4-manifolds with involutions, extending the powerful machinery of Seiberg–Witten theory to the Real setting.
• Resolution of the Existence Problem: It gives a precise cohomological characterization (via equivariant cohomology) for when a class can be represented by a Real surface, a question previously lacking a complete answer.
• Exotic Behavior: The results demonstrate that "Real" topology is distinct from "ordinary" topology. The existence of manifolds where the Real minimal genus is strictly larger than the ordinary minimal genus highlights the rigidity introduced by the involution.
• Applications to Real Algebraic Geometry: The motivation stems from real algebraic geometry (complexification of real curves). The results provide topological obstructions and genus bounds for real algebraic curves in complex surfaces with real structures.
• Technical Advancement: The development of equivariant blow-up formulas and the connected sum formula for Real Seiberg–Witten invariants are significant technical contributions that enable the calculation of these invariants on complex manifolds constructed from simpler building blocks.

In summary, Baraglia establishes that Real Seiberg–Witten theory provides effective lower bounds for the genus of Real embedded surfaces, revealing a gap between the minimal genus of Real surfaces and arbitrary surfaces in specific 4-manifolds.