🔢 mathematics

Krein's Formula for Conic Laplacians on Compact Riemann Surfaces

This paper establishes Krein's formula for self-adjoint extensions of conic Laplacians on compact Riemann surfaces by parametrizing them via Lagrangian subspaces of asymptotic boundary data, and utilizes the resulting trace identity to derive a comparison formula for positive-spectrum zeta determinants that accounts for negative eigenvalues and zero modes.

Original authors: Jia-Ming (Frank), Liou

Published 2026-06-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jia-Ming (Frank), Liou

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 a smooth, perfectly round drum (a Riemann surface). If you tap it, it vibrates in specific patterns, creating a unique musical tone. Mathematicians can calculate the "size" of this drum's sound using a special tool called a Zeta determinant. It's like a cosmic scorecard that sums up all the possible notes the drum can make.

Now, imagine poking a tiny, sharp hole in that drum. In math, this is called a conic singularity. The surface is no longer smooth; it has a pointy tip. When you tap this modified drum, the vibrations behave strangely near the tip. The math gets messy because the standard rules for smooth surfaces break down at that sharp point.

This paper by Jia-Ming (Frank) Liou is about figuring out exactly how to measure the "sound" (the Zeta determinant) of these pointy drums, even when we don't know exactly how the vibrations behave right at the tip.

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

1. The Problem: The "Fuzzy" Edge

When you have a smooth drum, the rules are clear. But at the sharp point (the singularity), the vibrations can act in many different ways. Some might die out quickly, others might grow wild.

The Math Reality: In the language of the paper, there are two main sets of rules for these vibrations: a "minimal" set (very strict, smooth behavior) and a "maximal" set (allows for wilder behavior near the point).
The Gap: The difference between these two sets isn't infinite; it's actually a small, finite list of specific "wiggle patterns" that can happen right at the tip. The author calls this the critical asymptotic boundary space. Think of it as a small control panel with a few knobs that determine how the drum behaves at the sharp point.

2. The Solution: The "Control Panel" (Lagrangian Subspaces)

Since there are a few specific ways the drum can behave at the tip, the author shows that we can describe every possible version of this "pointy drum" by choosing a specific setting on that control panel.

The Analogy: Imagine the control panel has a set of dials. To make the drum a valid, self-contained instrument (a "self-adjoint extension"), you must turn the dials to a specific configuration called a Lagrangian subspace.
The Result: Every valid way to fix the math at the sharp point corresponds to a unique setting on this panel. This turns a messy, infinite problem into a tidy, finite one.

3. The Bridge: Krein's Formula (The "Difference Engine")

The paper's main achievement is a new formula, called Krein's Formula.

The Analogy: Imagine you have two different settings on your control panel (two different ways of fixing the drum). You want to know the difference in their sound. Usually, calculating the difference between two complex systems is a nightmare.
The Magic: Liou's formula says, "Don't calculate the whole drum. Just look at the control panel." The difference between the two drums' sounds is entirely determined by the difference in their control panel settings and a special "translation map" (the Weyl function) that connects the panel to the sound.
Why it matters: This formula is incredibly efficient. It reduces a massive, infinite calculation to a simple calculation involving just a few numbers (the finite-dimensional boundary data).

4. The Grand Prize: Comparing the "Scores" (Zeta Determinants)

Finally, the author uses this efficient formula to compare the "scores" (Zeta determinants) of two different pointy drums.

The Challenge: When comparing the scores, you have to account for two tricky things:
1. Negative Notes: Some settings might produce "negative" vibrations (unstable modes) that the standard scorecard ignores.
2. Silence: Some settings might produce a "zero mode" (a vibration that doesn't vibrate at all).
The Result: The paper provides a precise recipe to compare the scores. It says:

"The ratio of the two scores equals a specific number derived from the control panel settings, adjusted by a correction factor that accounts for the negative notes and the silence."

Summary

In short, this paper takes a very difficult problem in geometry (measuring the properties of surfaces with sharp points) and solves it by:

Identifying a small, finite "control panel" that governs the behavior at the sharp point.
Creating a shortcut formula (Krein's Formula) that uses this panel to calculate the difference between any two versions of the surface.
Using that shortcut to write a perfect recipe for comparing the "musical scores" (Zeta determinants) of these surfaces, including corrections for any weird or silent vibrations.

It's a bit like realizing that to compare the sound of two broken violins, you don't need to analyze the wood grain of the whole instrument; you just need to look at how the two players are holding their bows and apply a simple rule to predict the difference in their music.

Technical Summary: Krein's Formula for Conic Laplacians on Compact Riemann Surfaces

Problem Statement
This paper addresses the spectral theory of conic Laplacians on compact Riemann surfaces with a finite set of conic singularities $S$ . Specifically, it investigates the self-adjoint extensions of the minimal conic Laplacian $\Delta_{min}$ defined on $L^2(X \setminus S)$ . In the presence of conic singularities, the domain of the maximal Laplacian $\Delta_{max}$ strictly contains the minimal domain $\Delta_{min}$ . The quotient space $D_{max}(\Delta)/D_{min}(\Delta)$ is finite-dimensional and carries a natural symplectic structure. The central problem is to characterize these self-adjoint extensions, derive a resolvent difference formula (Krein's formula) between any two such extensions, and utilize this to establish a comparison formula for their zeta determinants, accounting for contributions from negative eigenvalues and zero modes.

Methodology
The author employs a direct approach based on local Fourier expansions at the conic points, avoiding reliance on general cone pseudodifferential calculus or the abstract theory of boundary triples.

Boundary Symplectic Space: The paper constructs a finite-dimensional symplectic space $V_S = \bigoplus_{p \in S} V_p$ , where $V_p$ is the "critical asymptotic space" at a conic point $p$ with index $n_p$ . This space is spanned by critical asymptotic terms of the form $1, \log|z|, z^k, \bar{z}^k$ for $1 \le |k| \le n_p - 1$ . The quotient $D_{max}(\Delta)/D_{min}(\Delta)$ is identified with $V_S$ .
Green Form and Lagrangian Conditions: A skew-Hermitian sesquilinear form (the Green form) $q$ is defined on $D_{max}(\Delta)$ . It is shown that $q$ descends to a non-degenerate symplectic form $\omega$ on $V_S$ . Self-adjoint extensions of $\Delta_{min}$ are parametrized by Lagrangian subspaces $V \subset V_S$ (maximal isotropic subspaces).
Weyl Functions: For a given Lagrangian subspace $V$ and a complementary Lagrangian subspace $V^\#$ , the author defines a Weyl function $M_{V,V^\#}(\lambda): V^\# \to V$ . This operator acts as a finite-dimensional analogue of the Dirichlet-to-Neumann map, encoding the boundary data of formal solutions to $(\Delta - \lambda)u = 0$ .
Resolvent Difference: By analyzing the difference between resolvents $R_V(\lambda)$ and $R_W(\lambda)$ of two extensions, the author derives a Krein-type formula expressing this difference in terms of the Weyl function and the symplectic projections associated with the Lagrangian pairs.
Trace Identity and Zeta Determinants: The trace of the resolvent difference is computed, yielding a trace identity involving the logarithmic derivative of a boundary determinant function $d_{V,W}(\lambda)$ . This identity is then integrated to compare the positive-spectrum zeta determinants of different extensions.

Key Contributions and Results

Parametrization of Extensions: The paper rigorously establishes a one-to-one correspondence between self-adjoint extensions of the conic Laplacian and the Lagrangian Grassmannian of the boundary symplectic space $V_S$ .
Krein's Formula (Theorem 4.2): A precise formula is derived for the resolvent difference $R_V(\lambda) - R_W(\lambda)$ . The formula demonstrates that the difference is a finite-rank operator entirely determined by the boundary data and the Weyl function $M_{W,W^\#}(\lambda)$ .
Trace Identity (Theorem 4.4): The trace of the resolvent difference is shown to equal the logarithmic derivative of a determinant function constructed from the Weyl function:
$\text{Tr}(R_V(\lambda) - R_W(\lambda)) = \frac{d}{d\lambda} \log \det \left( C_V^\dagger M_{W,W^\#}(\lambda) + A_V^\dagger \right)$
Zeta Determinant Comparison (Theorem 5.2 & Corollary 5.3): The paper derives a comparison formula for the zeta determinants of two self-adjoint extensions $\Delta_V$ and $\Delta_W$ . Unlike standard gluing formulas, this result explicitly accounts for the "positive-spectrum" convention (omitting non-positive eigenvalues from the zeta function definition) and includes correction terms for:
- The negative spectrum (via the term $E_{V,W}(\mu)$ ).
- Zero modes (via the order of the pole of the boundary determinant at zero).
The final comparison formula at $\mu=0$ is:
$\frac{\text{Det}_\zeta(\Delta_V)}{\text{Det}_\zeta(\Delta_W)} = C_{V,W} (-1)^r \frac{d^{\text{red}}_{V,W}(0)}{\prod_{\lambda_\ell^V < 0} (\lambda_\ell^V)^{m(\lambda_\ell^V)} / \prod_{\lambda_\ell^W < 0} (\lambda_\ell^W)^{m(\lambda_\ell^W)}}$
where $r = \dim \ker \Delta_V - \dim \ker \Delta_W$ , and $d^{\text{red}}$ is the reduced boundary determinant.

Significance and Claims
The paper positions itself as complementary to the work of Hillairet and Kokotov, who utilized the S-matrix for Euclidean surfaces with conical singularities. The author claims that the presented approach offers a distinct perspective by formulating the resolvent and determinant comparison problems within a finite-dimensional boundary-symplectic framework.

Key claims regarding the significance of this work include:

Direct Normalization: By using local Fourier modes directly, the paper fixes the normalization of the Green form required for the subsequent trace identity, a detail often obscured in more abstract boundary triple formalisms.
Finite-Dimensional Reduction: The reduction of the infinite-dimensional spectral problem to a finite-dimensional symplectic space allows for explicit computation of resolvent differences and determinant ratios without needing the full S-matrix machinery.
Handling of Zero Modes and Negative Spectrum: The derived comparison formula is notable for its explicit inclusion of contributions from negative eigenvalues and zero modes, which are often omitted or treated separately in standard zeta determinant comparisons. This makes the formula applicable to general self-adjoint extensions that are not necessarily non-negative.

The paper concludes that this framework provides a robust tool for comparing spectral invariants of conic Laplacians under varying Lagrangian boundary conditions.