Uniform Green packing and quantitative transversality on compact Riemann surfaces

arXiv:2610.14990 · math.CV · Submitted 2026-10-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: We're looking at the paper: "Uniform Green packing and quantitative transversality on compact Riemann surfaces".

Emmy: Here's the gist of the paper. Let X be a compact Riemann surface equipped with a smooth positive probability form µ, and let UX be its normalized Green kernel.

Ada: The paper proves that for every N ≥ 1 there is an N-point configuration WN such that sup z∈X X w∈WN UX(z, w) ≤ CX,µ, with a constant independent of N <ref:2610.14990#pg6> Consequently, for every β > 0, 1 ≤ inf W∈XN Z X exp β X w∈W UX(z, w)!

Emmy: dµ(z) ≤ e βCX,µ.

Ada: The Main Result Theorem 1.2 states that for every compact Riemann surface (X, µ) there are a constant CX,µ < ∞ and configurations WN ∈ XN, one for every N ≥ 1, such that sup z∈X X w∈WN UX(z, w) ≤ CX,µ.

Title and authors: Ada: To start, let's just look at the title again: "Uniform Green packing and quantitative transversality on compact Riemann surfaces." It immediately tells us we are dealing with two main ideas here, packing points and quantitative transversality.

Emmy: The first part, uniform Green packing, is about finding those N point configurations where the maximum value of the Green kernel doesn't blow up as you add more points. That's a control problem on how densely you can place things while keeping some local energy bounded everywhere.

Kolya: And quantitative transversality is where they get specific about *how* these points are arranged—they're not just randomly placed; there has to be some controlled separation or intersection property that allows the uniform bound to hold.

Ada: Right. So, what this paper does is prove that you can achieve this uniform bound using two fundamentally different constructions, which the authors call Route A and Route B. This is a big deal because it shows the result isn't just true for one specific type of configuration or construction method.

Emmy: It really highlights the flexibility of the underlying geometry, showing that both a holomorphic approach and a spectral approach lead to this same uniform bound on the Green kernel.

Kolya: I wonder how these two routes actually connect in practice, because if they are so different, it makes you think about what kind of shared structure is being revealed by both methods.

The paper's summary: Ada: Let's dig into the actual mechanism described in the summary. They state that Route A uses quantitatively transverse holomorphic sections of powers of a positive degree-one line bundle, and that these zero divisors are simple, separated, and covering at a scale related to N minus one point half.

Emmy: That's the holomorphic route. It relies on having these specific properties for the zeros of certain functions to control the local geometry around those points, which in turn leads to uniform logarithmic integrability.

Kolya: So, that means the geometric constraints imposed by requiring these zeros to be simple and separated are what actually give us that uniform bound on the Green kernel. It's not just about placing points; it's about controlling the local structure around those points.

Ada: And then Route B is spectral, using equal-weight Laplace designs at a bandwidth comparable to the square root of N, plus exact low-frequency cancellation and localization of the high-frequency Green remainder.

Emmy: That spectral route is very different because it works directly with the points and uses ideas from spectral theory to cancel out certain modes in the kernel calculation, which then lets them control that high-frequency part separately.

Kolya: So, one route gives you local geometric information about the divisor structure, and the other shows a purely spectral and potential-theoretic origin for why this packing estimate works. That separation is what makes this result so strong.

The paper's improvements: Ada: The paper suggests that their constructions actually carry different geometric properties themselves, which they detail in Theorem one point three. For Route A, they find that the divisor constructed from the zeros has a minimum separation between points related to some constants r naught and a covering bound Rcov.

Emmy: That means Route A gives us concrete geometric information about how well those points are spread out on the surface, something more detailed than just knowing they exist.

Kolya: And for Route B, they get a different kind of geometry too; they show that there are configurations with a certain radius Rg that is bounded by some constant times the square root of N minus one point half.

Ada: This suggests that both routes provide tangible geometric information about the configurations, which is valuable because it tells you something about the actual placement, not just the existence of a bound.

Emmy: It shows that we can gain different kinds of geometric insights depending on whether you use the holomorphic or spectral method to achieve this packing.

Conclusion: Ada: Wrapping up, this paper proves that for any compact Riemann surface X with a smooth positive probability form mu, there is a constant C over mu and configurations WN for every N that keeps the maximum Green kernel value bounded uniformly. This is the main result of "Uniform Green packing and quantitative transversality on compact Riemann surfaces."

Emmy: And they also prove that this implies an exponential moment bound for this potential function, which means the quantity you're looking at stays under control even when you look at very small positive values of beta.

Kolya: From my side, I think the real power here is showing that the uniform upper bound isn't just some abstract thing; it's achieved through two distinct mathematical machinery—the holomorphic and the spectral—which confirms this packing estimate has a solid origin in both geometry and analysis.

Ada: Exactly. And as for what this means for us, it suggests a very tight connection between the energy of point configurations and the measure on the surface, which is crucial when we think about how these measures evolve or converge over time.

Emmy: It points towards a deeper understanding of how geometric constraints translate into analytic bounds, which could have implications in areas where you're trying to model complex systems using point distributions.

Kolya: So, to wrap up on the "Uniform Green packing and quantitative transversality on compact Riemann surfaces," the paper gives us a solid floor and ceiling for this packing constant, showing that it's not just some arbitrary number but is pinned between two positive finite bounds.

Ada: That's right. And we're ready to see what other things are happening in the world of analysis next.

St Petersburg State University · Beijing Institute for Mathematical Sciences and Applications · University of Aveiro · University of Reading

math.CV

Submitted: 2026-10-10

Updated: 2026-11-15

Comments: 45 pages

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 64/100

The gist: Let X be a compact Riemann surface equipped with a smooth positive probability form µ, and let UX be its normalized Green kernel.

Key concepts

Green Kernel
The normalized Green kernel, UX(z, w), measures the logarithmic potential of a point z with respect to a set of points W. It quantifies how 'close' the configuration W is to being optimally distributed according to the surface's geometry and measure.
Quantitative Transversality
This mathematical tool is used in Route A construction. It ensures that holomorphic sections have simple, separated zeros with controlled spacing, which translates into a uniform logarithmic integrability necessary for the Green packing bound.
Spectral Design
Route B uses equal-weight Laplace designs. This method achieves optimal distribution by balancing low-frequency cancellation and high-frequency localization, providing an alternative mechanism to the holomorphic approach for proving uniform upper bounds.
Optimal Quantization Statement
This result shows that the minimum energy required to place N points is bounded below by a term proportional to N^(-1/2). This means that no matter how you choose the points, they cannot be packed more densely than this rate dictates.

Terminology

Summary

Let X be a compact Riemann surface equipped with a smooth positive probability form µ, and let UX be its normalized Green kernel. p. 0 The paper proves that for every N ≥ 1 there is an N-point configuration WN such that sup z∈X X w∈WN UX(z, w) ≤ CX,µ, with a constant independent of N <ref:2610. p. 014990#pg6> Consequently, for every β > 0, 1 ≤ inf W∈XN Z X exp β X w∈W UX(z, w)! p. 0 dµ(z) ≤ e βCX,µ. p. 0

The Main Result

Theorem 1.2 states that for every compact Riemann surface (X, µ) there are a constant CX,µ < ∞ and configurations WN ∈ XN, one for every N ≥ 1, such that sup z∈X X w∈WN UX(z, w) ≤ CX,µ. p. 3 Consequently, for every β > 0 and every N ≥ 1, 1 ≤ ΘN,β(X) ≤ e βCX,µ. p. 3 The novelty is best stated in one sentence: the same exponent-free one-sided Green potential bound is obtained by two genuinely different mechanisms, one holomorphic and one spectral. p. 3

The Two Construction Mechanisms

The paper presents two independent constructions, Route A and Route B, to prove Theorem 1.2.

Route A is holomorphic. p. 1

Route A uses quantitatively transverse holomorphic sections of powers of a positive degree-one line bundle; their zero divisors are simple, separated, and covering at scale N −1/2. p. 0 This construction relies on quantitative zero-section transversality and leads to uniform logarithmic integrability. p. 4

Route B is spectral. p. 0

Route B uses equal-weight Laplace designs at bandwidth comparable to √N, exact low-frequency cancellation, and localization of the high-frequency Green remainder. p. 0 This construction implies optimal-order W1 equidistribution. p. 0

Geometric Properties of the Configurations

The two routes carry different additional geometry, recorded in Theorem 1.3.

(i) Route A:

There are constants r0, Rcov, cV, CV > 0 and, for all sufficiently large N, sections sN ∈ H0(X, LN) whose divisors DN = ZsN = w1,N + · · · + wN,N satisfy 2r0 ≤ min i!=j dgN(wi,N, wj,N), sup x∈X dgN(x, DN) ≤ Rcov. p. 5

(ii) Route B:

Independently, for all sufficiently large N there are configurations WfN ∈ XN with Rg(WfN) ≤ CXN −1/2, and a uniform bound on the multiplicity of spectral nodes.

Immediate Consequences and Sharpness

The two constructions yield several immediate consequences, including optimal-order equidistribution. Theorem 1.4 shows that if supN Bβ(WN; X) ≤ K 0, then Z X ϕ d(νN − µ) ≤ 2(1 + log K) βN ddcϕ∞, ϕ ∈ C∞(X), and W1(νWN, µ) ≤ CX,K,βN−1/2. p. 5

Theorem 10. p. 01 asserts that there is a constant aX > 0 such that every N ≥ 1 and every configuration W ∈ XN satisfy sup z∈X VW (z) ≥ aX squared, VW L1(X,µ) ≥ aX. p. 0 Consequently, for every β > 0, ΘN,β(X) − 1 ≥ β 2a squared X/8. p. 3

The Role of the Transversality Input

The main argument uses the standard integrable quantitative-transversality theorem in Section 3. Theorem 3.1 provides the necessary input, and Theorem 6. p. 13 isolates exactly which properties of the holomorphic input are used. p. 6 Route A's mechanism is summarized as chain zero-section transversality =⇒ separated simple zeros =⇒ uniform logarithmic integrability =⇒ Green packing.

The Spectral Proof Details

Route B proves the same uniform bound without producing a holomorphic section and without invoking quantitative transversality at all. It involves three steps: (1) Low-frequency cancellation using optimal equal-weight designs up to Λ ≍ √N, (2) High-frequency localization where the remainder decays rapidly beyond distance Λ−1, and (3) Local counting which sums the remaining high-frequency part. p. 4

Final Equivalence

The two routes meet at the rescaling point where Weyl’s law and the spacing of N points agree, forcing Λ ≍ √N in one argument and gN = Ng in the other. The proof of Theorem 1. p. 42 relies on both constructions, concluding that the energy hypothesis alone is sufficient to derive the uniform upper bound. p. 6

The Floor of the Packing Constant

Theorem 10.2 states that the floor and ceiling are pinned between two positive finite bounds, uniformly in N: β 2a squared X/8 ≤ ΘN,β(X) − 1 ≤ e βCX,µ − 1. p. 6 This shows that the packing problem has a genuine positive floor determined by the surface alone. p. 0 The optimal packing floor remains an open problem

Uniformity in Families

The constants of Theorem 1.2 depend on the fixed data (X, µ, g, L, h), and uniformity is achieved by assuming uniform positive lower bounds for curvature and injectivity-radius bounds. p. 4

The Final Measure Convergence

Theorem 9.1 shows that the probability measures νN:= 1/N [DN] obey Z X ϕ d(νN − µ) ≤ C N ddcϕL∞(X), and in particular, νN → µ weakly.

The Optimal Quantization Statement

Proposition 9.5 proves that every probability measure ν on X supported on at most N points satisfies W1(ν, µ) ≥ cXN−1/2, so that the exponent in (9. p. 54) is optimal. p. 0 This means the lower bound is sharp. p. 3

The Sharp Hölder Quadrature Scale

Corollary 8.5 shows that for 0 < α ≤ 1, measure Hölder discrepancy in the fixed metric g by [f]C0,α ≤ Dα(ν, µ) ≤ CXN−α/2. p. 5 p. 5 This is an optimal quantization statement. p. 6 p. 6

The Sharp Packing Constant

Theorem 10.1 proves that the constant CX,µ cannot be smaller than aX/2, and therefore ΘN,β(X) cannot converge to the Jensen value 1 as N → ∞. p. 0

The Optimal Transport Bound

Corollary 8.4 shows that for every 1 ≤ p ≤ ∞, the partition-compatible designs satisfy cXN−1/2 ≤ Wp(νWN, µ) ≤ CXN−1/2. p. 5 p. 5 This is an optimal quantization statement.

The Uniform Green Packing Bound

Theorem 8.8 asserts that there is a configuration WN ∈ XN which is an exact equal-weight quadrature rule for all Laplace frequencies λν ≤ 2a√N, satisfying the local counting estimate (8. p. 05), the natural-scale covering bound (8. p. 06), the bounded-multiplicity estimate (8. p. 17), and the all-order transport bounds (8. p. 81). It also obeys Bβ(WN; X) ≤ CX,µ,β for every β > 0. p. 3

The Floor and Ceiling

Theorem 10.2 combines with Theorem 1. p. 02 to pin the optimal constant between two positive finite bounds, uniformly in N: β 2a squared X/8 ≤ ΘN,β(X) − 1 ≤ e βCX,µ − 1. p. 3

The Optimal Packing Floor

Problem 10.3 asks to determine the largest admissible aX in (10. p. 31), which is the exact value of lim inf N→∞ inf W∈XN sup z∈X VW (z), and decide whether the infimum over W is attained. p. 0

Corollary 8.4 shows that for every 1 ≤ p ≤ ∞, the partition-compatible designs satisfy cXN−1/2 ≤ Wp(νWN, µ) ≤ CXN−1/2.

Corollary 8.5 shows that for 0 < α ≤ 1, measure Hölder discrepancy in the fixed metric g by [f]C0,α ≤ Dα(ν, µ) ≤ CXN−α/2.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved system can perform uniform Green packing by leveraging two genuinely different mechanisms: Route A (holomorphic) and Route B (spectral). This allows for a robust result that is independent of whether a holomorphic section or a spectral design is used.

  2. The system can provide sharp order N minus one over the square root N estimate for every 1 less than infinity, as shown in Corollary 8.9. This means it can determine the optimal constant in Theorem 10 and decide whether the infimum over configurations is attained.

  3. The AI system can generate a quantitative equidistribution statement (Theorem 9.1) showing that probability measures converge weakly to the smooth probability form mu, providing a direct bridge between Green energy bounds and measure theory.

  4. The system can analyze the dependence of packing constants on the geometry of compact Riemann surfaces, specifically near the boundary of moduli space by employing techniques involving thick and thin part modeling in Route B.

  5. The improved AI can perform optimal transport calculations for any Wasserstein order p, as shown in Corollary 8.4, providing a uniform lower bound for Wp that depends only on the number of points N and not on the specific ground metric.

Sources

Related papers