Finite-volume effects on smeared spectral densities

This paper derives a universal expression for the leading finite-volume effects on smeared vector-vector spectral densities using two distinct approaches, demonstrating that these effects are exponentially suppressed and governed by the pion form factor, thereby providing a framework to reliably estimate and control volume extrapolations in lattice QCD calculations.

The Gist

The Big Picture: Listening to a Room with Echoes

Imagine you are trying to understand the sound of a specific instrument (like a violin) playing in a room. In the real world (which physicists call "infinite volume"), the sound waves travel out forever, and you hear the pure, true tone of the instrument.

However, in the world of lattice Quantum Chromodynamics (QCD)—the computer simulations physicists use to study subatomic particles—the "room" is a tiny, invisible box with walls. Because the box is finite, the sound waves bounce off the walls and create echoes. These echoes distort the sound you hear, making it hard to tell what the instrument actually sounds like in the real world.

This paper is about figuring out exactly how those "echoes" (called finite-volume effects) change the sound, so scientists can mathematically remove them and hear the true tone.

The Specific Problem: Smearing the Sound

In this study, the scientists aren't just listening to a single note. They are looking at a "smeared spectral density."

• The Analogy: Imagine instead of hearing a single clear note, you are trying to hear a chord where the notes are slightly blurred or "smeared" together. In physics, this "smearing" is a mathematical tool used to smooth out noisy data so it's easier to analyze.
• The Goal: The researchers want to know: "If I take this blurred sound from a tiny box, how much does the box size change the result? And can I predict that change using a simple formula?"

The Two Ways They Solved It

The authors, Francesca A. Bresciani, Mattia Bruno, and Maxwell T. Hansen, used two different "maps" to solve this puzzle and found they led to the exact same destination.

1. The "Echo Chamber" Approach (Euclidean Correlators)
They started by looking at how the sound waves (mathematical correlations) behave inside the box. They knew that in a box, the waves bounce around. They took the math describing these bounces and applied a "smearing filter" to them.

• The Trick: They used a mathematical maneuver called a "Wick rotation." Think of this as turning a map upside down. Suddenly, a problem that looked like a messy, oscillating wave became a clean, decaying curve. This allowed them to see that the "echoes" die away very quickly as the box gets bigger, specifically following an exponential pattern (like a battery draining).

2. The "Resonance" Approach (Lellouch-Lüscher-Meyer)
They also started from a different angle: looking at the specific energy levels (resonances) that can exist inside the box. There is a famous rule in physics (the Lellouch-Lüscher-Meyer formalism) that connects the energy levels in a box to how particles scatter in the open world.

• The Result: By applying this rule to the "blurred" sound, they derived the exact same formula as the first method.

The Main Discovery: The "Universal Formula"

The most important finding is a universal formula (Equation 25 in the paper) that predicts how much the "echoes" distort the result.

• What it depends on: The formula says the distortion depends on two main things:

  1. The Pion Form Factor: This is like the "fingerprint" of the particle interaction. It tells us how the particles (pions) behave when they hit each other.
  2. The Smearing Kernel: This is the specific "blur" filter the scientists chose to use.

• The "Exponential" Good News: The paper proves that for a certain class of these filters, the error caused by the box size shrinks exponentially as the box gets bigger.

  • Analogy: If you double the size of the room, the echo doesn't just get half as loud; it gets much, much quieter, almost vanishing. This means that if you have a box that is "big enough," you can trust the data very highly.

Why This Matters (According to the Paper)

The paper explains that this formula is a tool for control.

• The "Scaling Regime": The authors show that you can use this formula to find the "sweet spot" where the box is large enough that the leading "echo" is the only thing that matters. Once you are in this zone, you can reliably predict what the result would be in an infinite room without needing to simulate an impossibly huge box.
• Verification: They tested their formula with different models of particle interactions (like the "Gounaris-Sakurai" model, which describes a specific particle resonance called the rho meson). They found that the formula works consistently across these different models.

Summary

In short, this paper provides a mathematical recipe to calculate how much a tiny, computer-simulated "box" distorts the measurement of particle interactions.

By using two different mathematical paths, they proved that for certain types of data smoothing, the distortion follows a predictable, rapidly fading pattern based on how particles interact (the pion form factor). This allows scientists to take data from small computer boxes and confidently correct it to understand how the universe works in the real, infinite world.

Technical Summary

Problem Statement
Lattice Quantum Chromodynamics (QCD) calculations are performed in finite spatial volumes with periodic boundary conditions, yielding discrete samples of Euclidean correlators. While the Osterwalder-Schrader theorem guarantees the analytic continuation to real time in principle, this is practically intractable with noisy numerical data. Consequently, the community increasingly relies on reconstructing "smeared" spectral densities from Euclidean correlators. A critical systematic uncertainty in this approach is the finite-volume correction ($\delta \hat{\rho}_{L,\kappa}$) required to extrapolate results to the infinite-volume limit ($L \to \infty$). While finite-volume effects on stable hadron masses and Euclidean correlators are known to be exponentially suppressed, the specific behavior of these effects for smeared spectral densities (proportional to the hadronic $R$-ratio) has not been universally characterized. The challenge is to derive a reliable, universal expression for these corrections to control the $L \to \infty$ extrapolation, particularly identifying a scaling regime where the leading terms dominate.

Methodology
The authors derive the leading finite-volume effects on the smeared vector-vector spectral density, $\hat{\rho}_{L,\kappa}(\alpha)$, using two distinct and complementary theoretical frameworks:

1. Euclidean Correlator Approach: Building on previous work regarding finite-volume effects on the Euclidean two-point function $G_L(\tau)$, the authors start with the expression for the difference $\delta G_L(\tau)$ derived from low-energy effective theory. They apply the linear reconstruction coefficients $c_i(\alpha)$ (used to define the smearing kernel $\hat{\kappa}$) directly to this finite-volume difference. A crucial step involves a Wick rotation of the integration contour in the complex plane. This transforms the integrand from a form that decays exponentially in the volume $L$ and oscillates in Euclidean time $\tau$, to a form that decays in $\tau$ and oscillates in $L$. This rotation allows the finite-volume correction to be expressed in terms of the timelike pion form factor.

2. Lellouch-Lüscher-Meyer (LLM) Formalism: The authors alternatively start from the spectral decomposition of the correlator in terms of finite-volume energy eigenstates. Using the Lüscher quantization condition to relate finite-volume energies to infinite-volume scattering phase shifts, and the LLM formalism to relate finite-volume matrix elements to infinite-volume amplitudes (specifically the timelike pion form factor), they construct the smeared spectral density. By applying the Poisson summation formula to the sum over discrete energy levels, they derive an expression for the finite-volume correction.

Key Contributions and Results
The primary result of the paper is a universal expression for the leading finite-volume correction, $\delta \hat{\rho}_{L,\kappa}(\alpha)$, which is shown to be identical when derived via both methods. The correction is expressed as:
$$\delta \hat{\rho}_{L,\kappa}(\alpha) \approx f_\kappa(L, \alpha)$$
where $f_\kappa(L, \alpha)$ is an integral over the infinite-volume spectral density weighted by the smearing kernel and a function $\Delta(\omega, L)$ that encodes the finite-volume effects.

Key technical findings include:

• Exponential Suppression: For a certain class of smearing kernels, the finite-volume effects are exponentially suppressed in $L$, generally with power-law prefactors. This mirrors the behavior of Euclidean correlators and stable hadron masses.
• Universality: The leading correction is expressed universally in terms of the infinite-volume timelike pion form factor, $F(s)$, and the scattering phase shift $\delta_{\pi\pi}(p)$.
• Convergence and Scaling: The authors demonstrate that the series over Poisson modes (labeled by $n$) converges rapidly for sufficiently large $L$ and specific smearing widths. They identify that the asymptotic behavior is dominated by the nearest singularities in the complex plane, which can be either kinematic poles or poles associated with the smearing kernel itself.
• Numerical Validation: Using three models for the form factor (non-interacting, scattering volume, and Gounaris-Sakurai) and two smearing kernels (exponential and Cauchy), the authors numerically evaluate the corrections. They show that the effective mass of the finite-volume correction follows the expected $1/L e^{-mL}$ scaling in the asymptotic regime, with deviations at smaller $L$ attributed to power-law prefactors and higher-order terms.

Significance and Claims
The paper claims that its derivation provides a robust theoretical framework for controlling the $L \to \infty$ extrapolation of smeared spectral densities in lattice QCD. Specifically:

• It allows for the definition of a "scaling regime" where the finite-volume effects are dominated by the leading terms in a large-$L$ expansion, making them reliably estimable.
• The result enables lattice practitioners to either subtract leading finite-volume effects or include them explicitly in $L$-dependent fit functions, thereby improving the precision of phenomenological predictions (such as the hadronic vacuum polarization contribution to the muon $g-2$).
• The work establishes a direct theoretical link between the spacelike description of finite-volume effects (via Euclidean correlators) and the finite-volume matrix-element formalism (LLM), resolving previous questions about the consistency of different estimation methods.
• While focused on the vector channel, the authors posit that the derivation defines a general framework applicable to other spectral densities, including those involving multi-particle states (e.g., $D^0$-$\bar{D}^0$ mixing or three-particle states), provided the relevant Lellouch-Lüscher-type relations are known.

The authors remain modest regarding the scope, noting that the utility of the result depends on the specific smearing kernel chosen and that the representation is most useful when the leading terms in the expansion dominate. They do not claim to solve the inverse problem of spectral reconstruction but rather to provide the necessary tool for managing the finite-volume systematic error once a reconstruction method is chosen.