🔢 mathematics

Motivic height zeta function rationality and Kudla-Millson modularity for elliptic surfaces

This paper establishes the rationality and Hodge-theoretic stabilization of the trivial lattice specialization of a trivariate motivic height zeta function for minimal elliptic surfaces over function fields, while simultaneously demonstrating that the distribution of new Mordell-Weil sections is governed by transcendental modular forms via the Kudla-Millson theta correspondence.

Original authors: Jun-Yong Park

Published 2026-04-01

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Jun-Yong Park

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 a detective trying to understand the hidden patterns of a vast, magical city called Elliptic Surfaces. This city is built on a foundation of numbers and shapes, and its inhabitants are "elliptic curves"—mathematical objects that look like twisted loops but behave like complex machines.

Your goal is to count these inhabitants and understand how they grow as the city gets bigger. The paper you're asking about is a report on two very different ways these inhabitants organize themselves: one group is predictable and local, while the other is mysterious and global.

Here is the story of the paper, broken down into simple concepts.

1. The City and the Map

First, let's set the scene. The mathematician, Jun-Yong Park, is studying a specific type of city built over a field of numbers (like a giant grid).

The Inhabitants: These are "minimal elliptic curves." Think of them as unique, perfectly tuned engines.
The Height: Every engine has a "height," which is like a measure of its complexity or size. The paper looks at engines of height 1, height 2, height 3, and so on, forever.
The Goal: Park wants to create a "Zeta Function." In math, this is like a master inventory list. It's a giant formula that, if you plug in a number, tells you exactly how many engines exist at every level of complexity.

2. The Two Types of "Ranks" (The Shioda-Tate Formula)

The paper splits the complexity of these engines into two distinct parts, using a famous rule called the Shioda-Tate Formula. Imagine an engine has two main features:

The Trivial Lattice (The Local Blueprint):
- What it is: This is the "skeleton" of the engine. It's built from the standard parts that appear whenever the engine breaks down in specific, predictable ways (like a flat tire or a loose bolt).
- The Metaphor: Think of this like traffic lights. If you look at a specific intersection, you know exactly how many cars will stop based on the light color. It's a local rule. If you know the condition of the road at one spot, you know the traffic there.
- The Discovery: Park proves that if you only count these "skeleton" parts, the master inventory list is Rational.
- What "Rational" means here: It means the pattern is simple and predictable. It's like a recipe: "For every 12 steps of complexity, the number of engines grows by this exact factor." You can write this pattern down as a simple fraction. Even though the city is huge, the "skeleton" part follows a strict, local rule.
The Mordell-Weil Rank (The Global Mystery):
- What it is: This is the "extra" complexity. These are the special, unique features that appear out of nowhere, connecting distant parts of the engine.
- The Metaphor: Think of this like secret underground tunnels connecting different parts of the city. You can't see them just by looking at one intersection. To know if a tunnel exists, you have to understand the entire city at once.
- The Discovery: This is where things get wild. Park proves that if you try to count these "extra" features, the pattern is Transcendental.
- What "Transcendental" means here: It means the pattern is too complex to be written as a simple recipe. It's not just a fraction; it's a chaotic, beautiful, infinite dance. The distribution of these extra features is governed by something called Modular Forms (specifically, the Kudla-Millson correspondence).

3. The Two Main Results (The "Aha!" Moments)

Result A: The Local Part is Simple (Approximate Rationality)

Park shows that the "skeleton" part of the inventory (the Trivial Lattice) can be predicted with incredible precision.

The Analogy: Imagine you are counting the number of bricks in a wall. Even if the wall is miles long, you know that every 10 feet, there are exactly 50 bricks. You can predict the total count perfectly.
The Catch: The math is slightly tricky because the "bricks" (the engine parts) interact in a non-linear way (like a jigsaw puzzle where the pieces don't fit perfectly until the very end). Park uses a technique called Motivic Discriminant Stabilization (a fancy way of saying "we fix the messy edges by looking at the big picture") to show that, for all practical purposes, the count is a simple, rational fraction.

Result B: The Global Part is Chaotic (Modularity and Transcendence)

When Park looks at the "extra" features (the Mordell-Weil rank), the rules change completely.

The Analogy: Imagine trying to predict the number of people who will win the lottery. You can't just look at the ticket price; you have to look at the entire universe of possibilities. The pattern of winners follows a Modular Form.
The Metaphor: Modular forms are like perfectly symmetrical snowflakes or musical chords that repeat in a complex, self-similar way. They are beautiful, but they cannot be described by a simple "if-then" recipe.
The Conclusion: The paper proves that the way these "extra" features appear is transcendental. This means the pattern of how the "rank" jumps up and down is so complex that no finite list of algebraic rules can ever fully capture it. It's a "global" phenomenon that resists being broken down into local pieces.

4. Why Does This Matter?

This paper is a bridge between two worlds:

The World of Local Rules: Where things are predictable, rational, and can be counted like bricks.
The World of Global Magic: Where things are governed by deep, symmetrical, and chaotic laws (Modular Forms) that we can see but not fully simplify.

The Big Takeaway:
The paper tells us that while the "skeleton" of our mathematical universe (the trivial lattice) follows simple, local laws, the "soul" of the universe (the Mordell-Weil rank) is governed by a deep, global magic that is too complex to be tamed by simple formulas. It's a beautiful reminder that in mathematics, the more you look at the whole picture, the more complex and wonderful it becomes.

Summary in One Sentence

The paper proves that the "local" parts of elliptic curves follow a simple, predictable recipe, but their "global" secrets are governed by a complex, magical symmetry that cannot be reduced to a simple formula.

1. Problem Statement and Context

The paper investigates the arithmetic and geometric statistics of minimal elliptic curves over the function field $K = k(t)$ (where $k$ is a perfect field of characteristic $\neq 2, 3$ ). Specifically, it focuses on the motivic height zeta function, which counts rational points on the moduli stack of elliptic curves, refined by two key invariants derived from the associated elliptic surface $S$ :

Trivial Lattice Rank ( $T(S)$ ): The rank of the lattice generated by the zero section, fiber class, and components of reducible fibers.
Mordell–Weil Rank ( $\text{rk}(E/K)$ ): The rank of the group of rational sections.

The central problem is to understand the generating series (zeta function) that weights these counts by $u^{T(S)} v^{\text{rk}(E/K)}$ . The author aims to determine whether this series is rational (or approximable by rational functions) and to contrast the behavior of the trivial lattice (local data) against the Mordell–Weil rank (global data).

2. Methodology

The paper employs a synthesis of motivic integration, moduli theory, and automorphic forms:

Motivic Height Moduli: The author utilizes the framework of Bejleri–Park–Satriano [BPS22], viewing minimal elliptic curves as rational points on the weighted projective stack $\mathbb{P}(4,6) \simeq \mathcal{M}_{1,1}$ . The height corresponds to the degree of the discriminant.
Stratification by Kodaira Types: The moduli space is stratified by the configuration of singular fibers (Kodaira types). The trivial lattice rank $T(S)$ is constant on each stratum, allowing for a decomposition of the zeta function.
Motivic Transversality and Stabilization:
- For the trivial lattice, the author adapts the Vakil–Wood method [VW15]. Since the discriminant $\Delta = 4a_4^3 + 27a_6^2$ is a nonlinear function of the Weierstrass coefficients, prescribing root multiplicities at many points is not independent. The author uses a truncated inclusion–exclusion principle in the dimensionally completed Grothendieck ring ( $\hat{\mathcal{M}}_L$ ) to handle the "invisible" fibers (those contributing 0 to $T(S)$ ) and prove approximate rationality.
- For the Mordell–Weil rank, the author utilizes the Kudla–Millson theta correspondence [KM90] and its extension by Garcia [Gar18] to period domains. This connects the distribution of new algebraic cycles (Noether–Lefschetz loci) to modular forms.
Hodge Theory: Over $k=\mathbb{C}$ , the dimensional filtration is translated into Hodge number stabilization using mixed Hodge theory weight bounds.

3. Key Contributions and Results

A. Rationality of the Trivial Lattice Specialization

The paper proves that the specialization of the zeta function where $v=1$ (focusing only on the trivial lattice rank $T(S)$ ) is approximately rational.

Theorem 1.5: Let $s = t^{1/12}$ $s = t^{1/12}$ . For any fixed power $u^m$ $u^{m}$ , the coefficient of $u^m$ $u^{m}$ in the zeta function $Z_{\text{Triv}}(u; t)$ $Z_{Triv} (u; t)$ lies in the closure of rational functions $\hat{\mathcal{M}}_L(s)$ $\hat{M}_{L} (s)$ within the dimensional completion.
- Mechanism: The series is approximated by rational functions $R_{m,N}(s)$ with an error term that vanishes in the dimensional filtration as the truncation depth $N \to \infty$ .
- Structure: The approximation arises from an Euler product over local fiber types.
  - Isotrivial loci ( $j=0, 1728$ ): The discriminant is a pure power of a single coefficient, yielding exact rational Euler products.
  - Non-isotrivial locus: The additive fibers are handled by linear independence of vanishing conditions. The multiplicative fibers (roots of $\Delta$ ) are handled via the Vakil–Wood stabilization method.
Corollary 1.6 (Hodge Stabilization): Over $k=\mathbb{C}$ , the virtual Hodge numbers of the moduli spaces stabilize bidegree-wise for large $n$ , implying the generating function of these numbers is rational in $s$ .

B. Transcendence of the Mordell–Weil Distribution

In sharp contrast to the trivial lattice, the distribution of the Mordell–Weil rank is shown to be transcendental.

Theorem 1.7: The generating series of Noether–Lefschetz classes (which correspond to jumps in the Mordell–Weil rank) on the period domain $D_n$ $D_{n}$ associated with the lattice $\Lambda_n = \langle F, O \rangle^\perp$ $Λ_{n} = ⟨ F, O ⟩^{⊥}$ is a holomorphic modular form of weight $6n-2$ $6 n - 2$ for $SL_2(\mathbb{Z})$ $S L_{2} (Z)$ .
- Modularity: The series $\sum [H_d] q^d$ (where $H_d$ is the Heegner locus where a vector of square $-2d$ becomes algebraic) is a modular form.
- Transcendence: For any non-trivial test class $\alpha$ , the scalar-valued series $\phi_{n,\alpha}(q)$ is transcendental over $\mathbb{C}(q)$ .
- Implication: This proves that the pattern of Mordell–Weil rank jumps cannot be captured by a finite Euler product or local data. The rank is a "genuinely global" phenomenon governed by the geometry of the period domain.

C. Structural Distinction

The paper provides a rigorous structural explanation for the Shioda–Tate formula ( $\rho(S) = T(S) + \text{rk}(E/K)$ ):

$T(S)$ (Trivial Lattice): Determined by local reduction data (fiber configurations). This locality allows for Euler product structures and approximate rationality.
$\text{rk}(E/K)$ (Mordell–Weil): Determined by global algebraic cycles appearing in the orthogonal complement of the trivial lattice. This is governed by the theta correspondence and modular forms, resisting local-to-global factorization.

4. Significance

Motivic Arithmetic: The work extends the theory of motivic zeta functions to include lattice rank refinements, demonstrating that "local" arithmetic invariants (trivial lattice) behave differently from "global" ones (Mordell–Weil rank) in the motivic setting.
Modularity of Noether–Lefschetz Loci: It establishes a direct link between the arithmetic of elliptic surfaces over function fields and the theory of modular forms via the Kudla–Millson correspondence, generalizing previous results on K3 surfaces.
Transcendence Results: It provides a concrete example where the distribution of rational points (rank jumps) is transcendental, contrasting with the rationality often found in counting problems over finite fields or for local invariants.
Methodological Innovation: The adaptation of the Vakil–Wood stabilization method to the nonlinear discriminant of elliptic curves ( $\Delta = 4a_4^3 + 27a_6^2$ ) offers a new tool for handling non-linear conditions in motivic integration.

5. Conclusion

Jun-Yong Park's paper successfully decomposes the complexity of elliptic surface moduli into two distinct behaviors: the trivial lattice, which is locally determined and approximable by rational functions, and the Mordell–Weil rank, which is globally determined and governed by modular forms. This dichotomy provides deep insight into the arithmetic geometry of elliptic surfaces over function fields and highlights the power of combining motivic integration with automorphic forms.