🔢 mathematics

Kodaira-Iitaka dimension and multiplicity: an analytic perspective

This paper establishes analytic formulas for the Kodaira-Iitaka dimension and multiplicity of graded linear series using the intersection theory of plurisubharmonic envelopes, deriving refined pointwise and metric-level results that lead to a new non-pluripolar numerical invariant dominating the classical dimension while remaining bounded by existing numerical versions.

Original authors: Siarhei Finski

Published 2026-03-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Siarhei Finski

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 an architect trying to understand the shape and size of a mysterious, invisible building. You can't see the whole thing at once, but you have a collection of blueprints (sections) and a set of tools (metrics) to measure them. This paper is about figuring out exactly how big this building is and what its "dimension" really means, using a mix of geometry, calculus, and a bit of magic called "pluripotential theory."

Here is a breakdown of the paper's main ideas using simple analogies:

1. The Problem: Measuring the Invisible Building

In mathematics, we often study "linear series." Think of these as a growing collection of blueprints for a building.

The Building: A complex geometric shape (a manifold).
The Blueprints: Mathematical functions (sections) that describe the building.
The Goal: We want to know two things:
1. Kodaira-Iitaka Dimension: How "big" is the building? Is it a flat sheet (2D), a solid block (3D), or something in between?
2. Multiplicity: How "dense" are the blueprints? If you stack them up, how much volume do they take up as the collection grows infinitely large?

Traditionally, mathematicians calculated these by looking at the algebra of the blueprints. This paper asks: Can we measure these things using the "shape" of the space itself, like measuring the temperature or pressure in a room?

2. The Tool: The "Plurisubharmonic Envelope" (The Inflatable Bubble)

The authors introduce a concept called the plurisubharmonic (psh) envelope.

The Analogy: Imagine you have a messy pile of blueprints scattered on the floor. You want to cover them with a giant, smooth, invisible balloon (the envelope) that hugs the blueprints as tightly as possible from above.
The Magic: This balloon isn't just a random shape; it's the "smoothest" possible shape that still respects the constraints of the blueprints.
The Discovery: The authors found that the dimension and density of your blueprint collection are directly encoded in the "curvature" of this balloon. If the balloon is flat in some directions, the building is lower-dimensional. If it curves everywhere, it's full-dimensional.

3. The Two New Ways to Measure

The paper offers two refined ways to measure this, moving from a "point-by-point" view to a "global volume" view.

A. The Pointwise View (The "Crowd Density" Meter)

The Concept: They look at Bergman kernels. Think of these as a "crowd density meter" that tells you how many blueprints pass through a specific point in space.
The Surprise: Usually, if you have a perfect, complete set of blueprints, this meter settles down to a smooth, predictable pattern as you add more blueprints.
The Twist: The authors found that for partial or incomplete sets of blueprints, this meter can go crazy! It might oscillate or diverge, never settling on a single pattern.
The Lesson: This happens because the "envelope balloon" isn't always smooth enough to be perfectly measured in every direction. It's like trying to measure the wind speed in a room with a sudden, jagged draft; the reading jumps around.

B. The Metric View (The "Volume of the Suitcase")

The Concept: Imagine you have a suitcase (the space of all possible blueprints). You want to know how much space the "best" blueprints take up.
The Method: They compare the volume of a suitcase filled with blueprints measured under two different "rulers" (metrics).
The Formula: They discovered a beautiful formula: The ratio of these volumes is directly linked to the energy stored in the difference between the two "envelope balloons."
The Metaphor: It's like saying, "The difference in how much space my clothes take up in a summer suitcase vs. a winter suitcase is exactly determined by the difference in the 'heat' (energy) of the fabric."

4. The New "Non-Pluripolar" Dimension

The authors introduce a new, stricter way to measure the dimension of a line bundle (the building).

The Old Way: Sometimes, the "envelope balloon" has tiny, invisible holes (pluripolar sets) that the old math ignores.
The New Way: They define a Non-Pluripolar Dimension. This is a dimension that refuses to ignore those tiny holes.
The Result: They proved that this new dimension is always larger than or equal to the old dimension.
- Analogy: If the old dimension says "This building is 3D," the new dimension might say, "Actually, because of these tiny, hidden cracks, it's effectively 4D in a very specific mathematical sense."
- They also showed this new dimension sits between the old one and other proposed "numerical" dimensions, acting as a perfect middle ground.

5. The "Birational" Exception (When Things Work Perfectly)

There is one special case where everything behaves beautifully: when the blueprint collection is birational.

The Analogy: This is like having a set of blueprints that perfectly maps the building without any missing pieces or weird overlaps.
The Result: In this case, the "crowd density meter" (Bergman kernel) does settle down smoothly. The "envelope balloon" is well-behaved, and the energy formulas work perfectly.
The Counterexample: The authors even built a specific, simple example (using two copies of a projective line) where the collection is not birational, and the meter goes haywire. This proves that their new, more complex formulas are necessary.

Summary

This paper is a bridge between algebra (counting blueprints) and analysis (measuring shapes and energies).

Old View: Count the blueprints to find the dimension.
New View: Inflate a mathematical balloon over the blueprints. The shape and energy of that balloon tell you the dimension and density.
Key Insight: Sometimes the balloon is jagged, causing measurements to fluctuate. But by understanding why it fluctuates (using "non-pluripolar" math), we get a more accurate, robust definition of the building's size.

It's a bit like realizing that to measure the size of a cloud, you can't just count the water droplets; you have to understand the pressure and temperature of the air surrounding it.

1. Problem Statement

The paper addresses the fundamental problem of characterizing the Kodaira-Iitaka dimension ( $\kappa(W)$ ) and the multiplicity (asymptotic volume, $vol_{\kappa}(W)$ ) of a graded linear series $W = \bigoplus_{k=0}^\infty W_k$ associated with a holomorphic line bundle $L$ on a complex projective manifold $X$ .

While these invariants are classically defined algebraically (via the growth rate of $\dim W_k$ ) or via intersection theory on resolutions of singularities, Finski seeks to establish purely analytic formulas for them. Specifically, the paper aims to:

Express $\kappa(W)$ and $vol_{\kappa}(W)$ using the plurisubharmonic (psh) envelope associated with the linear series and intersection theory of currents.
Refine these formulas at two distinct levels:
- Metric Level: Relating the asymptotic ratio of volumes of unit balls (defined by sup-norms) to a relative energy functional.
- Pointwise Level: Analyzing the weak convergence of partial Bergman kernels.
Investigate the behavior of these invariants for non-big line bundles and non-birational linear series, where classical results (often restricted to big line bundles) may fail.

2. Methodology

The author employs tools from pluripotential theory, complex geometry, and asymptotic analysis. Key methodological components include:

Fubini-Study Maps and Envelopes: The paper utilizes the Fubini-Study operator to associate singular metrics to norms on $W_k$ . The central object is the psh envelope $P[W, h_L]$ , defined as the limit of Fubini-Study metrics associated with sup-norms on $W_k$ . This generalizes the classical psh envelope used for complete linear series.
Intersection Theory of Currents: The author defines non-pluripolar products of currents $c_1(L, P[W, h_L])^i \wedge \omega^{n-i}$ to measure the "size" of the linear series analytically.
Fujita-type Approximation: The proofs rely heavily on the approximation of arbitrary linear series by subrings generated by sections of ample line bundles (via the Lazarsfeld-Mustatǎ/Kaveh-Khovanskii theorems). This allows reducing general cases to situations where the geometry is better understood (e.g., semi-ample bundles).
Relative Monge-Ampère Energy: The paper introduces and utilizes a relative Monge-Ampère $\kappa$ -energy, a functional measuring the difference between envelopes of two metrics, restricted to the dimension $\kappa(W)$ .
Bernstein-Markov Measures: To study pointwise convergence, the author considers measures satisfying the Bernstein-Markov property, which ensures that $L^2$ norms (induced by the measure) are asymptotically equivalent to sup-norms.

3. Key Contributions and Results

A. Analytic Formulas for Dimension and Multiplicity

The paper establishes two main theorems expressing the algebraic invariants in terms of analytic data:

Theorem 1.1 (Dimension): The Kodaira-Iitaka dimension $\kappa(W)$ is the maximal integer $i$ such that the non-pluripolar intersection product of the envelope's curvature current is non-zero:
$\kappa(W) = \max \{ i \in \{0, \dots, n\} : c_1(L, P[W, h_L])^i \wedge \omega^{n-i} \neq 0 \}$
Theorem 1.3 (Multiplicity): The multiplicity is given by the integral of the top-degree non-pluripolar product of the envelope, normalized by a fiber integral:
$vol_{\kappa}(W) = \frac{1}{\phi^*[\omega]^{n-\kappa(W)}} \int_X c_1(L, P[W, h_L])^{\kappa(W)} \wedge \omega^{n-\kappa(W)}$
Here, $\phi$ represents the rational map associated with the linear series.

B. Metric Refinement: Volume Ratios and Energy

The paper refines the multiplicity formula by comparing volumes of unit balls defined by different metrics.

Theorem 1.7: The asymptotic ratio of the volumes of unit balls $B_k[W, K, h_L^0]$ and $B_k[W, K, h_L^1]$ (defined by sup-norms over a non-pluripolar set $K$ ) converges to the difference of their relative Monge-Ampère $\kappa$ -energies:
$\lim_{k \to \infty} \frac{1}{k^{\kappa(W)+1}} \log \frac{v(B_k[W, K, h_L^0])}{v(B_k[W, K, h_L^1])} = E_\kappa(P[W, K, h_L^0]) - E_\kappa(P[W, K, h_L^1])$
This generalizes the Berman-Boucksom result for big line bundles to arbitrary graded linear series.

C. Pointwise Refinement: Partial Bergman Kernels

The paper investigates the weak convergence of partial Bergman kernels $\mu_k = \frac{1}{k^{\kappa(W)}} B_k(x,x) d\mu$ .

Counterexample (Section 5.1): The author constructs an explicit example showing that for non-birational linear series, the sequence of partial Bergman kernels does not necessarily converge weakly. This failure is linked to the non-differentiability of the relative Monge-Ampère $\kappa$ -energy.
Theorem 1.9 (Partial Convergence): While global weak convergence may fail, the paper proves that if the linear series factors through a rational map $\rho: X \to Z$ , then the pushforward of the measures $\rho_*(\mu_k)$ converges weakly to an equilibrium measure $\mu_{eq}$ on $Z$ .
$\rho_*(\mu_k) \rightharpoonup \mu_{eq}[W, Z, K, h_L]$
If $W$ is birational, this recovers the standard weak convergence on $X$ .

D. Non-Pluripolar Numerical Invariants

The paper introduces a non-pluripolar numerical Kodaira-Iitaka dimension ( $\kappa_{np}(L)$ ) and multiplicity, defined via the minimal singularity current $T_{min}$ .

Inequality Chain: The author proves the following hierarchy of invariants for any effective line bundle $L$ :
$\kappa(L) \leq \kappa_{np}(L) \leq \kappa_{mov}(L)$
where $\kappa_{mov}$ is the movable intersection dimension. The paper demonstrates that the first inequality can be strict, proving that the classical Kodaira-Iitaka dimension is not a numerical invariant (it depends on the specific linear series, not just the cohomology class).

E. Integral and Analytic Closures

In Section 6, the paper relates the integral closure of a linear series to an analytic closure defined via the singularity type of the psh envelope.

Theorem 2.1: For the integral closure $\overline{W}$ , the Kodaira-Iitaka dimension remains unchanged ( $\kappa(\overline{W}) = \kappa(W)$ ), while the multiplicity scales by the generic degree of the map between the associated images.

4. Significance

Unification: The paper unifies algebraic geometry (Kodaira-Iitaka dimension) with complex analysis (plurisubharmonic envelopes and Monge-Ampère currents), providing a robust analytic framework for studying linear series.
Extension beyond Big Bundles: It significantly extends the scope of known results (previously mostly limited to big line bundles and complete series) to arbitrary graded linear series and non-big line bundles.
Refinement of Convergence: By identifying the precise conditions under which partial Bergman kernels converge (requiring a factorization through a base space), the paper clarifies the limitations of previous convergence theorems and introduces the correct notion of convergence for general series.
New Invariants: The introduction of the non-pluripolar numerical dimension provides a new tool to distinguish between the numerical properties of a line bundle and the specific properties of its linear series, resolving questions about the numerical nature of the Kodaira-Iitaka dimension.

In summary, Finski's work provides a comprehensive analytic toolkit for understanding the asymptotic geometry of linear series, bridging the gap between algebraic definitions and analytic potentials, and revealing subtle behaviors in the non-big and non-birational regimes.