Residues of a tropical zeta function for convex domains

✨

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 have a perfectly smooth, convex cookie (like a round cookie or an oval one) sitting on a giant, infinite grid of graph paper. This grid represents the "lattice" of numbers.

Now, imagine you are a tiny ant walking on this cookie. You want to know how far you are from the edge, but you can't just walk in a straight line like a human would. Instead, you can only walk in directions that align perfectly with the grid lines (up, down, left, right, or diagonal steps that hit grid intersections).

This paper introduces a new mathematical tool called a Tropical Zeta Function to study this specific way of measuring distance on the grid. Here is the breakdown of what the authors discovered, using simple analogies.

1. The "Tropical Distance" (The Ant's View)

In normal geometry, distance is measured with a ruler (Euclidean distance). But in this "Tropical" world, the distance is defined by the "supporting lines" of the grid.

The Analogy: Imagine the cookie is a fortress. The "Tropical Distance" isn't how far you are from the wall in a straight line; it's how many "grid-steps" you have to take to get out, following the specific rules of the grid's geometry.
The authors define a function, $\rho$ , that measures this special distance from any point inside the cookie to the edge.

2. The "Zeta Function" (The Sound of the Cookie)

The authors take this distance function and turn it into a "Zeta Function."

The Analogy: Think of the cookie as a musical instrument. If you pluck it, it makes a sound. The "Zeta Function" is like the sheet music for that sound. It doesn't just tell you the volume; it tells you the frequencies (poles) that make up the sound.
In math, these "frequencies" (called poles) reveal hidden secrets about the shape. Where the music gets loud or changes pitch tells you about the geometry of the cookie.

3. The Big Discovery: Two Different "Songs"

The paper shows that the "song" the cookie sings depends entirely on how smooth its edge is.

Case A: The Polygonal Cookie (The Bumpy Edge)

If your cookie has straight edges and sharp corners (like a square or a triangle), the "song" has a very obvious, loud note at a specific pitch ( $s=1$ ).

What it means: This loud note simply counts the "lattice perimeter." It's like the cookie is shouting, "I have 12 grid-steps around my edge!"
The Takeaway: For bumpy shapes, the math is straightforward; it just counts the grid steps along the edge.

Case B: The Smooth Cookie (The Curved Edge)

If your cookie is perfectly smooth (like a circle or an ellipse) with no flat edges, the "loud note" at $s=1$ disappears.

The Surprise: Instead, a new, more subtle note appears at a different pitch ( $s=2/3$ ).
The Secret: This new note doesn't measure the "Euclidean" length (the length you'd measure with a ruler). Instead, it measures something called Equiaffine Perimeter.
The Analogy: Imagine stretching a rubber sheet over the cookie. The "Equiaffine Perimeter" is a measure of how the cookie resists being squashed or stretched in a specific way. It's a "shape invariant" that stays the same even if you stretch the grid paper.
The Result: The authors proved that the "loudness" of this new note is directly proportional to this special curved length.

4. The "Magic Trick" (Turning Inside Out)

How did they figure this out? They performed a mathematical magic trick.

The Trick: They showed that the complex calculation happening inside the cookie (the volume integral) is actually exactly the same as a calculation happening on the edge (a boundary series).
The Metaphor: It's like realizing that the sound of a drum (the inside) is determined entirely by the tension of the drumhead's rim (the edge). You don't need to look inside the drum to know its sound; you just need to analyze the rim.
They turned the problem into a "boundary Dirichlet series," which is a sum of numbers related to the grid directions along the edge.

5. The "Parabolic" Connection (The Universal Shape)

The authors found that the simplest shape to understand this math is a parabola (the curve of a thrown ball).

The Connection: When they did the math for a parabolic shape, the result turned out to be a famous number series known as Witten's SU(3) Zeta Function.
Why it matters: This connects a simple geometry problem to deep physics (Quantum Field Theory) and number theory. It suggests that the parabola is the "fundamental building block" for understanding how these grid-based distances work on smooth curves.

6. The "Wave Front" (The Shrinking Cookie)

Finally, they looked at what happens if you slowly eat the cookie from the outside in, keeping the "grid-distance" constant. This creates a shrinking "wave front."

The Prediction: They used their new math to predict exactly how fast the "lattice perimeter" of this shrinking wave grows as it gets very small.
The Result: For a smooth cookie, the perimeter grows at a rate related to $t^{1/3}$ (where $t$ is the time). This is a very specific, non-intuitive speed that only appears because of the grid's influence on the curve.

Summary

This paper is about finding a new way to "listen" to the shape of a convex object using the rules of a grid.

If the shape is bumpy, the math counts the grid steps.
If the shape is smooth, the math reveals a hidden "affine" length that is invisible to normal rulers but visible to the grid.
The authors proved that the "music" of the grid (the Zeta function) changes its tune from a simple count to a complex geometric invariant ( $s=2/3$ ) when the shape becomes smooth, and they figured out exactly how to calculate that new tune.

It's a beautiful bridge between Number Theory (counting grid points), Geometry (measuring curves), and Physics (using tools from quantum theory).

1. Problem Statement and Motivation

The paper addresses the analytic properties of lattice point counting in dilates of compact convex domains $\Omega \subset \mathbb{R}^n$ . While the classical Gauss circle problem and its generalizations focus on the error term in counting lattice points $N(R\Omega) = \text{Vol}(\Omega)R^n + E(R)$ , the authors introduce a new tool to study the fine structure of this error term: the Tropical Zeta Function.

The central problem is to understand the singularities (poles) of this zeta function and relate them to the geometric and arithmetic properties of the boundary $\partial \Omega$ . Specifically, the authors investigate how the transition from polygonal domains (with rational slopes) to smooth strictly convex domains (with non-vanishing curvature) alters the analytic structure of the zeta function, shifting the dominant singularity from a lattice-perimeter dependent pole to an affine-geometric dependent pole.

2. Definitions and Setup

The Tropical Distance Function:
For a compact convex domain $\Omega$ , the authors define a "tropical distance-to-the-boundary" function $\rho_\Omega(x)$ for $x \in \Omega^\circ$ :
$\rho_\Omega(x) = \min_{u \in \mathbb{Z}^n_{\text{prim}}} (\langle u, x \rangle - h_\Omega(u))$
where $h_\Omega(u) = \min_{y \in \Omega} \langle u, y \rangle$ is the lower support function, and $\mathbb{Z}^n_{\text{prim}}$ denotes primitive lattice vectors. This function is piecewise linear, concave, and vanishes on the boundary. It is invariant under $SL(n, \mathbb{Z})$ and translations.

The Tropical Zeta Function:
The tropical zeta function is defined as the integral:
$Z_\Omega(s) = \int_\Omega \rho_\Omega(x)^{s-n} \, dV$
Initially defined for $\Re(s)$ sufficiently large, the authors study its meromorphic continuation.

Mellin Representation:
Using the "layer-cake" formula, $Z_\Omega(s)$ is shown to be the Mellin transform of the lattice perimeter profile of the tropical wave fronts $\Omega_t = \{x \in \Omega : \rho_\Omega(x) \geq t\}$ :
$Z_\Omega(s) = \int_0^{m_\Omega} t^{s-n} P_\Omega(t) \, dt$
where $P_\Omega(t)$ is the lattice surface volume (perimeter in 2D) of the level set $\Omega_t$ .

3. Methodology

The paper employs a multi-faceted approach combining tropical geometry, symplectic geometry, analytic number theory, and convex geometry:

Minimal Models and Symplectic Geometry:
The authors associate $\Omega$ with a "minimal model" $\hat{\Omega}$ , a rational-slope polytope obtained by "cutting" $\Omega$ from a larger polytope via unimodular corner cuts. This process is interpreted as a sequence of symplectic blow-ups. The difference between the zeta function of $\Omega$ and its minimal model is expressed via a boundary Dirichlet series.
Reduction to Boundary Series:
A key technical result (Theorem 1) establishes an exact identity in dimension 2:
$s(s-1)Z_\Omega(s) = -F_{\partial \Omega}(s) + H_{\hat{\Omega}}(s)$
Here, $F_{\partial \Omega}(s)$ is a boundary Dirichlet series built from "support defects" (areas of triangles cut during the corner-cutting process) and Farey neighbors, while $H_{\hat{\Omega}}(s)$ is a holomorphic term determined by the minimal model. This reduces the analytic study of the interior integral to the arithmetic study of the boundary series.
Farey Intervals and Hata Coefficients:
The boundary series is reinterpreted using Farey intervals and Hata's basis for continuous functions. The terms of the series correspond to weighted sums over primitive lattice directions. The authors utilize Legendre duality to transform the geometric support defects into a form involving the second derivative of the boundary curve.
Analytic Number Theory:
To analyze the boundary series near the critical point, the authors employ:
- Equidistribution of reduced residues: Using Fejér kernels to approximate sums over coprime integers.
- Incomplete Kloosterman sums: Utilizing Weil bounds to control error terms.
- Tauberian Theorems: To convert the analytic properties of the zeta function (poles) into asymptotic estimates for the lattice perimeter of the wave fronts as $t \to 0$ .

4. Key Results

A. The Polygonal Case (Rational Slopes)

If $\Omega$ is a polygon with rational slopes:

$Z_\Omega(s)$ extends meromorphically to the whole complex plane.
The rightmost pole is at $s=1$ .
The residue at $s=1$ is the lattice perimeter of $\partial \Omega$ .
The residue at $s=0$ relates to the negative self-intersection of the canonical class of the associated toric surface.

B. The Smooth Strictly Convex Case (Main Result)

Let $\Omega \subset \mathbb{R}^2$ be a compact convex domain with $C^3$ boundary and everywhere non-vanishing curvature.

Disappearance of the $s=1$ pole: The pole at $s=1$ vanishes because the boundary contains no straight segments.
New Leading Singularity: The leading singularity moves to $s = 2/3$ .
Residue Formula: The residue at $s=2/3$ is explicitly proportional to the equiaffine arc length of the boundary:
$\text{Res}_{s=2/3} Z_\Omega(s) = \frac{3^{5/2}}{2^{5/3}\pi^3 \Gamma(1/3)^3} \cdot \text{Length}_{\text{equiaffine}}(\partial \Omega)$
where $\text{Length}_{\text{equiaffine}}(\partial \Omega) = \int_{\partial \Omega} \kappa^{1/3} ds$ .

C. The Parabolic Model

The authors explicitly compute the zeta function for a domain bounded by a parabolic arc.

The boundary series reduces to the primitive Mordell–Tornheim series, which is related to Witten's $SU(3)$ zeta function.
This model confirms the pole at $s=2/3$ and the specific transcendental constants appearing in the general residue formula.
This domain is also identified as the limit shape for random lattice polygons in a square (concentration of measure).

D. Asymptotics of Tropical Wave Fronts

Using a Tauberian argument on the boundary series, the authors derive the short-time asymptotic for the lattice perimeter of the inward wave fronts $\Omega_t$ as $t \to 0^+$ :
$P_\Omega(t) \sim C \cdot \text{Length}_{\text{equiaffine}}(\partial \Omega) \cdot t^{1/3}$
This connects the arithmetic $SL(2, \mathbb{Z})$ symmetry of the zeta function to the affine $SL(2, \mathbb{R})$ geometry of the boundary.

5. Significance and Implications

Bridge between Arithmetic and Geometry: The paper demonstrates that the singularities of a zeta function constructed from discrete lattice data (primitive support directions) encode continuous affine geometric invariants (equiaffine length) in the smooth limit.
New Tool for Lattice Point Problems: The tropical zeta function offers a new perspective on the error term in lattice point counting. While the classical error term is bounded by $O(R^{2/3})$ for smooth convex domains (van der Corput), the pole at $s=2/3$ in the tropical zeta function suggests a deep structural reason for this exponent, linking it to the equiaffine geometry of the boundary.
Symplectic and Toric Connections: The work provides a rigorous tropical interpretation of symplectic blow-ups and minimal models, linking the combinatorics of Farey sequences to the topology of toric surfaces.
Universality: The result suggests that for any smooth strictly convex domain, the "first nontrivial" singularity is universal (governed by $s=2/3$ and equiaffine length), distinguishing it from the polygonal case where the singularity is determined by the lattice perimeter.

In summary, the paper establishes that the tropical zeta function serves as a "spectral" tool for convex geometry, where the spectrum (poles) transitions from reflecting discrete lattice properties (perimeter) to continuous affine properties (equiaffine length) as the domain smooths out.