Source-linked AI summary
W physics at the LHC with FEWZ 2.1
Ryan Gavin, Ye Li, Frank Petriello, Seth Quackenbush
TL;DR
The paper addresses the need for precise, practical predictions for W and γ*/Z production at hadron colliders. It updates FEWZ to provide NNLO calculations, new observables, broader PDF support, and faster integration, then demonstrates W-physics studies and LHC comparisons. The update achieves major speed improvements, shows excellent agreement with early LHC data, and demonstrates that direct fiducial comparisons preserve sensitivity to PDF differences.
Problem
Precise W-production predictions are needed for PDF constraints, precision measurements, and searches because LHC statistical errors are negligible while systematic errors remain.
Method
FEWZ 2.1 extends FEWZ 2.0 with NNLO W-production calculations, new observables, LHAPDF integration, arbitrary cuts, and faster parallelized integration.
Results
FEWZ 2.1 delivers an additional order-of-magnitude speed gain, achieves NNLO precision with standard acceptance cuts in hours on one machine, and agrees excellently with early LHC data.
Takeaways & Limitations
Direct comparisons of fiducial measurements with theory avoid extrapolation-induced experimental uncertainty and retain PDF-discrimination power.
Takeaways & Limitations
The example W results use standard acceptance cuts: MT > 40 GeV, ETmiss > 25 GeV, pTl > 20 GeV, and |ηl| < 2.5.
Abstract
from arXiv · showhide
We present an updated version of the FEWZ (Fully Exclusive W and Z production) code for the calculation of W and gamma*/Z production at next-to-next-to-leading order in the strong coupling. Several new features and observables are introduced, and an order-of-magnitude speed improvement over the performance of FEWZ 2.0 is demonstrated. New phenomenological results for W production and comparisons with LHC data are presented, and used to illustrate the range of physics studies possible with the features of FEWZ 2.1. We demonstrate with an example the importance of directly comparing fiducial-region measurements with theoretical predictions, rather than first extrapolating them to the full phase space.
1 Introduction
W production is both a precision channel and an important background at the LHC, making accurate fiducial predictions and efficient NNLO tools valuable. FEWZ 2.1 extends FEWZ 2.0 to W physics while adding observables, PDF support, and substantially faster integration.
- W production constrains quark and antiquark PDFs through charge-asymmetry measurements and supports precise W-boson mass determinations.
- W production and W-plus-jets processes form major backgrounds to searches for new W′ bosons, supersymmetry, and other missing-energy signatures.
- Few inverse femtobarns of LHC data have made statistical errors negligible, leaving experimental and theoretical systematic errors at the few-percent level or smaller.
- FEWZ 2.1 applies NNLO QCD predictions to W production, supports arbitrary cuts, incorporates LHAPDF PDF sets, and introduces several new observables.
- An additional order-of-magnitude speed gain improves the practicality of FEWZ calculations, while fiducial comparisons with LHC data and PDF predictions are demonstrated.
2 New features
FEWZ 2.1 extends FEWZ 2.0 with W-production support, new observables and histogram controls, and substantially faster integration. These features enable broader W-physics studies while preserving access to consistent uncertainty propagation and improved computational performance.
- Program extension: FEWZ 2.1 adds W-production support alongside the existing γ*/Z calculation through dedicated fewzw and fewzz executables sharing most source code.Both executables use similar basic and histogram input files, with sector differences reflecting W and γ*/Z diagrams and symmetry factors at NNLO.
- New observables: Two new observables—beam thrust and transverse mass—expand the available W-physics studies.Beam thrust supports central-jet-veto studies without a jet algorithm, while transverse mass provides a W-mass-sensitive distribution without requiring the neutrino’s longitudinal momentum.
- Output features: Cumulative histograms, user-defined variable bin sizes, and an asymmetry operator provide more flexible analysis and output control.The asymmetry operator combines W+ and W− runs, with statistical and PDF errors propagated consistently.
- Performance improvements: Sector recomposition and revised sector combinations reduce integrand variance and improve the mapping of integration variables to parton phase space.The reworked combinations reduce the time required to reach target precision; FEWZ 2.1 contains 133 sectors for fewzz and additional sectors for the W calculation.
- Performance improvements: An order of magnitude less time is required to reach a target integration precision than with FEWZ 2.0 under the benchmark setup.The comparison uses the same cuts, histograms, PDF sets, and benchmark machine, apart from the two new histograms.
3 Example results
FEWZ 2.1 applies standard fiducial cuts to NNLO W-production predictions, enabling distributions, PDF comparisons, and direct comparisons with LHC measurements. The examples show distinct PDF behavior, radiation-sensitive observables, and why fiducial measurements retain greater discriminating power than extrapolated inclusive results.
- The analysis imposes MT > 40 GeV, ETmiss > 25 GeV, pT^l > 20 GeV, and |η^l| < 2.5 as standard acceptance cuts.
- Fiducial W^+ and W^- cross sections and their ratio are evaluated with HERA, MSTW, and NNPDF NNLO PDF fits, including 68% C.L. PDF errors.Parametric uncertainties are omitted, while scale uncertainties are described as subdominant and integration errors are below 0.1%.
- NNPDF and HERA predictions agree with the ATLAS fiducial ratio within one sigma, whereas MSTW differs by over two sigma from the measurement's central value.
- The inclusive-ratio uncertainty is twice as large as the fiducial-ratio uncertainty after extrapolating measurements to the full phase space, and all three PDF sets then agree within one sigma.
- W^± transverse-mass and rapidity distributions display Jacobian features, while higher-order recoil populates 40 GeV < MT < 50 GeV.W^+ enhancement near |Y| ∼2 reflects the stronger up-quark PDF peaking than the down-quark distribution at x ∼0.1.
- Beam-thrust and cumulative leading-jet-pT distributions illustrate new observables; their behavior reflects the smaller CFαs/π radiation factor in W production.No significant differences among the three PDF sets are found for either observable.
- W^+/W^- distribution ratios expose PDF-set separation at central lepton rapidity and connect larger W^+ rapidity fractions to smaller W^+ lepton transverse momenta at large η.
4 Conclusions
FEWZ 2.1 extends FEWZ 2.0 with new observables and supports NNLO W-physics studies, including comparisons with LHC data. The update also demonstrates distributions and ratios relevant to phenomenological analyses.
- FEWZ 2.1 adds new observables and extends FEWZ 2.0 for NNLO W-physics studies at hadron colliders.
- Figure 3 presents bin-integrated W− beam-thrust and cumulative W+ leading-jet pT distributions.
- Figure 4 presents W+ over W− production ratios versus lepton pseudorapidity and lepton pT.
- Comparisons with early LHC data indicate excellent agreement between data and FEWZ 2.1 theory predictions.