VV Resummation To NNLO+NNLL At the LHC

This paper presents threshold-resummed predictions for vector boson pair (WW and ZZ) production at the LHC at NNLO+NNLL accuracy, demonstrating that the inclusion of resummed corrections reduces scale uncertainties and adds a few percent to the fixed-order results.

The Gist

Imagine you are trying to predict exactly how many people will show up to a massive, chaotic concert (the Large Hadron Collider, or LHC) and what they will be doing. Specifically, you are interested in two very specific types of VIP guests: pairs of "Vector Bosons" (let's call them ZZ and WW). These particles are crucial for understanding how the universe works, but they are also the "background noise" that hides other, even more exciting discoveries.

To make accurate predictions, physicists use complex math called Quantum Chromodynamics (QCD). However, calculating these predictions is like trying to predict the weather: the more you look, the more variables you find, and the more uncertain you become.

Here is what this paper does, broken down into simple concepts:

1. The Problem: The "Static" on the Radio

When physicists calculate how often these particle pairs are made, they usually do it in steps, like adding layers to a cake.

• The First Layer (Fixed Order): This is the basic recipe. It's good, but it has "static" or fuzziness.
• The Issue: As you get closer to the "edge" of the energy spectrum (the threshold), the math gets messy. It's like trying to listen to a radio station right at the edge of its broadcast range; the signal gets weak and filled with static. In physics terms, these are called "threshold logarithms." If you ignore them, your prediction is a bit off, and you aren't sure how off it is (this is called "scale uncertainty").

2. The Solution: The "Noise-Canceling Headphones"

The authors of this paper developed a new method called Resummation.

• The Analogy: Imagine you are trying to hear a whisper in a noisy room. You could try to calculate every single sound wave in the room (the "Fixed Order" method), but it's overwhelming. Instead, you put on noise-canceling headphones (Resummation). These headphones specifically target and cancel out the background static (the soft-gluon effects) so you can hear the whisper clearly.
• The Upgrade: They didn't just use basic headphones; they built Next-to-Next-to-Leading Logarithmic (NNLL) headphones. This is the highest level of precision currently available for this type of problem. They then combined this with their best existing "Fixed Order" calculations (NNLO) to get the ultimate prediction: NNLO+NNLL.

3. What They Found: Sharper Images and Less Guesswork

When they applied this new "noise-canceling" math to the LHC data, two big things happened:

A. The Numbers Changed (A Little Bit)
The new, sharper predictions showed that the actual number of particle pairs produced is slightly higher (by a few percent) than the old, fuzzy predictions.

• Analogy: It's like realizing your old map said a city was 100 miles away, but with a GPS upgrade, you realize it's actually 103 miles away. It's a small change, but for a race car driver (or a particle physicist), that 3 miles matters.

B. The "Fuzziness" Disappeared (Reduced Uncertainty)
This is the most important part. In the old calculations, the physicists had to guess at certain "knobs" (called scales) to get their numbers, and changing the knobs changed the result by about 4%. That's a lot of guesswork.

• With the new Resummation method, that guesswork dropped to about 2.8%.
• Analogy: Imagine you are trying to hit a target with a bow and arrow. The old method was like shooting in the dark; your arrows landed in a big, messy circle (high uncertainty). The new method is like turning on a spotlight and using a laser sight; your arrows now land in a tight, precise group (low uncertainty).

4. The "Dynamic" vs. "Static" Choice

The paper also tested different ways to set the "central scale" (the baseline for the calculation).

• Static Scale: Like setting your car's cruise control to a fixed 60 mph regardless of the road.
• Dynamic Scale: Like an adaptive cruise control that speeds up and slows down based on the traffic ahead.
• The Result: The "Dynamic" approach (adjusting the scale based on the energy of the collision) gave much more stable and reliable results than the fixed approach.

The Bottom Line

This paper is a major upgrade to the "GPS" physicists use to navigate the LHC. By using advanced math to cancel out the background noise (resummation), they have made the predictions for how particles behave sharper, more precise, and less full of guesswork.

This is vital because if we want to find new physics (like dark matter or new particles) hidden inside the data, we first need to know exactly what the "old" physics (the Standard Model) looks like. If our map of the old physics is blurry, we might miss the new discoveries. This paper makes the map crystal clear.

Technical Summary

Here is a detailed technical summary of the paper "VV Resummation To NNLO+NNLL At the LHC" by Banerjee et al.

1. Problem Statement

Vector boson pair ($VV$, where $V=W, Z$) production at hadron colliders is a fundamental process for testing the Standard Model (SM), constraining anomalous gauge couplings, and serving as a background for Higgs and Beyond-the-Standard-Model (BSM) searches. While Next-to-Leading Order (NLO) and Next-to-Next-to-Leading Order (NNLO) QCD corrections have been computed, achieving percent-level theoretical precision requires addressing large logarithmic corrections that arise near the partonic threshold.

In the threshold limit ($z \to 1$, where $z = Q^2/\hat{s}$), the available phase space for additional parton radiation is restricted, causing soft-gluon emissions to dominate. These emissions generate large logarithmic terms ($\ln^k(1-z)$) that degrade the convergence of the perturbative series. To achieve the necessary precision for current and future LHC data (specifically at $\sqrt{S} = 13.6$ TeV), these threshold logarithms must be resummed to high accuracy and matched with fixed-order calculations.

2. Methodology

The authors perform a threshold resummation calculation for on-shell $WW$ and $ZZ$ production, matching the results to NNLO fixed-order QCD predictions to achieve NNLO+NNLL (Next-to-Next-to-Leading Logarithmic) accuracy.

• Theoretical Framework:

  • The calculation utilizes the factorization of the hadronic cross-section into parton distribution functions (PDFs) and a partonic cross-section.
  • The partonic cross-section is separated into a Singular-Virtual (SV) component (universal, containing threshold logarithms and $\delta(1-z)$ terms) and a Regular component.
  • Resummation is performed in Mellin space ($N$-space), where convolutions become simple products. The SV component is exponentiated as $\hat{\sigma}_N \propto \exp(\Psi_{sv})$, where $\Psi_{sv}$ resums the large logarithms $\ln N$.
  • The exponent $\Psi_{sv}$ is expanded up to NNLL accuracy: $\Psi_{sv} = \ln(\bar{N})g_1(\bar{N}) + g_2(\bar{N}) + a_s g_3(\bar{N}) + \dots$.

• Computational Implementation:

  • Coefficients: The process-dependent coefficients ($g_{01}, g_{02}$) required for the NNLL accuracy were computed using in-house codes. The two-loop amplitudes necessary for the $g_{02}$ coefficient were constructed using the public package VVamp.
  • Matching: The resummed result in $z$-space is obtained via inverse Mellin transformation. To avoid double-counting, the resummed result is matched to the fixed-order NNLO result using the standard "add-subtract" scheme:
    $$\sigma_{\text{matched}} = \sigma_{\text{fixed-order}} + (\sigma_{\text{resummed}} - \sigma_{\text{truncated}})$$
  • Prescription: The "Minimal Prescription" is used for the inverse Mellin transform to avoid the Landau pole in the complex $N$-plane.
  • Scales and PDFs:
    • $ZZ$ Production: Calculated in the 5-Flavor Scheme (5FS) using MSHT20 PDF sets.
    • $WW$ Production: Calculated in the 4-Flavor Scheme (4FS) using NNPDF30 sets to avoid resonant top-quark contamination from bottom-quark emissions.
    • Scales: Central scales are chosen as the invariant mass ($Q$), with variations between $Q/2$ and $2Q$.

3. Key Contributions

• First NNLO+NNLL Matched Results: The paper presents the first threshold resummation for $WW$ and $ZZ$ production matched to NNLO fixed-order results at the LHC. Previous works had only matched NNLL to NLO.
• High-Precision Amplitudes: The inclusion of two-loop amplitudes (via VVamp) allows for the calculation of the $g_{02}$ coefficient, which is essential for NNLL accuracy.
• Comprehensive Uncertainty Analysis: The study provides a detailed breakdown of scale uncertainties (7-point variation) for both differential (invariant mass) and total cross-section observables, comparing fixed central scales against dynamic scales.

4. Key Results

• K-Factors and Corrections:
  • The NNLL resummation adds a few percent to the fixed-order NNLO results.
  • For $WW$ production, the NNLO K-factor ($K_{20}$) ranges from 1.38 to 1.81, while the NNLO+NNLL K-factor ($R_{20}$) ranges from 1.39 to 1.86.
  • The resummation corrections are most pronounced in the high invariant mass region ($Q > 1000$ GeV), where threshold logarithms are largest.
• Reduction of Scale Uncertainties (Differential):
  • Resummation significantly improves perturbative stability in the high-$Q$ region.
  • At $Q = 1200$ GeV:
    • $ZZ$ Uncertainty: Reduced from 4.06% (NNLO) to 2.88% (NNLO+NNLL).
    • $WW$ Uncertainty: Reduced from 3.74% (NNLO) to 2.72% (NNLO+NNLL).
• Total Cross-Section Behavior:
  • Unlike the differential distribution, the total cross-section (dominated by the low-$Q$ region) shows a slight increase in scale uncertainty upon adding resummation. This is expected as the resummed corrections are larger in the low-$Q$ region where the fixed-order series is already well-behaved.
• Scale Choice Optimization:
  • Central Scale: A central scale choice of $\mu_0 = 2Q$ yields smaller uncertainties compared to $\mu_0 = Q$ or $Q/2$.
  • Dynamic vs. Fixed: Using a dynamic scale ($\mu_0 = Q$) results in significantly more stable predictions (lower uncertainties) compared to a fixed scale ($\mu_0 = m_W$) for both fixed-order and resummed calculations.

5. Significance

This work represents a critical step forward in the theoretical precision of vector boson pair production at the LHC. By achieving NNLO+NNLL accuracy, the authors have:

1. Reduced Theoretical Uncertainties: Specifically in the high-mass tail, which is crucial for BSM searches where new physics signals often appear.
2. Validated Perturbative Stability: Demonstrated that resummation stabilizes the perturbative expansion against renormalization and factorization scale variations.
3. Provided Benchmarks: The results serve as essential benchmarks for experimental analyses at the LHC (Run 3 and High-Luminosity LHC), enabling more stringent constraints on anomalous gauge couplings and more accurate background estimations for Higgs and new physics searches.