Further Reduction of the PDF Uncertainty in the High-Mass Drell-Yan Spectrum Utilizing Neutral and Charged Current Inputs

TL;DR

This paper extends a previous neutral-current PDF-reduction strategy to include charged-current inputs, using ePump to create boutique PDFs based on three-dimensional NC and two-dimensional CCDY differential information. By updating to the CT18 global fits and incorporating CC data, the authors demonstrate substantial reductions in high-mass Drell-Yan PDF uncertainties, enabling more robust background estimates for Z' and W' searches at the LHC. The results show that combining NC and CC inputs reduces uncertainties by factors up to several times, with explicit improvements in the relevant observables $d\sigma/dm_{\ell\ell}$ and $d\sigma/dm_T$, and even partial Run 3 data offer meaningful gains. The work highlights a practical path for leveraging future HL-LHC data to suppress PDF-related systematics in high-mass BSM searches and calls for adoption of these boutique PDFs by PDF-fitting groups.

Abstract

Uncertainties in the parametrization of Parton Distribution Functions are a serious limiting systematic uncertainty in Large Hadron Collider searches for Beyond the Standard Model physics. This is especially true for measurements at high scales induced by quark and anti-quark collisions, where Drell-Yan continuum backgrounds are dominant. In Phys. Rev. D99, 054004 (2019) we presented a unique strategy for improving uncertainties using neutral current Drell-Yan backgrounds and here we update that strategy and include charged current Drell-Yan final states in the program and demonstrate significant improvements. Through a judicious selection of measurable kinematical quantities can reduce the assigned systematic PDF uncertainties by significant factors in limit-setting or discovery for neutral and charged, high mass Intermediate Vector Bosons. This approach will be take advantage of the huge statistical precision of future High Luminosity, Large Hadron Collider Standard Model datasets and could also improve uncertainties in the high statistics results from LHC Run 3.

Figures (19)

• Figure 1: Lowest order diagrams for (a) the production of neutral IVBs, $\gamma/Z^0$ bosons in proton--proton collisions and (b)the production of charged IVBs, $W^\pm$.
• Figure 2: Historical distributions of the dilepton invariant mass in a hypothetical $q\overline{q} \to \gamma^{*}/Z' \to \ell^{+}\ell^{-}$ reaction in of both ATLAS (a) ref:ATLAS13TeV:ZPrime2017 and (b) ref:ATLAS13TeV:ZPrime2019 and CMS (c) ref:CMS13TeV:ZPrime2018 and (d) ref:CMS13TeV:ZPrime2021. ATLAS (b) has chosen to use a data-driven approach, essentially extrapolating the low mass DY continuum into the search region.
• Figure 3: Most recent distributions of the transverse mass in a hypothetical $q\overline{q} \to W^\pm \to \ell^{\pm}\nu_{\ell}$ reaction in of both (a) ATLAS and (b) CMS.
• Figure 4: The CT18 quark PDFs evaluated at a scales of (a) $Q=2$ GeV and (b) $Q=3$ TeV, while their uncertainties are shown in (c) and (d).
• Figure 5: The various experimental inputs and their relative importance in the overall fits for (a) $u_v$ and (b) $\bar{u}$ are shown in terms of the sensitivity quantity $|L_2|$. See the text for explanation.