The impact of new neutrino DIS and Drell-Yan data on large-x parton distributions

TL;DR

This paper assesses how new neutrino DIS (NuTeV, Chorus) and Drell–Yan (E-866) data constrain large-$x$ parton distributions within global NLO MS-bar fits. By comparing fits with and without heavy-target corrections, it highlights tensions between NuTeV and E-866, particularly in the $d/u$ ratio, and demonstrates that combining NuTeV and E-866 can flatten $d$ at large $x$ while inducing tension with other datasets. The work underscores the critical role of nuclear corrections and the potential value of heavy-target–independent measurements (e.g., E-866) to obtain a robust, model-insensitive determination of $d/u$. The Appendix provides the deuteron correction parameterization used to adjust deuteron-target data. The findings imply that reducing large-$x$ nuclear corrections could reconcile datasets and improve global PDF determinations.

Abstract

New data sets have recently become available for neutrino and antineutrino deep inelastic scattering on nuclear targets and for inclusive dimuon production in pp pd interactions. These data sets are sensitive to different combinations of parton distribution functions in the large-x region and, therefore, provide different constraints when incorporated into global parton distribution function fits. We compare and contrast the effects of these new data on parton distribution fits, with special emphasis on the effects at large x. The effects of the use of nuclear targets in the neutrino and antineutrino data sets are also investigated.

Figures (13)

• Figure 1: Comparison between the reference fit and the E-866 data.
• Figure 2: Comparison between the reference fit and the Chorus and NuTeV neutrino cross section data.
• Figure 3: Comparison of the $d/u$ ratios from the reference fit and the CTEQ6.1M PDFs at $Q^2=10 {\rm \ GeV}^2$.
• Figure 4: Same as Fig. \ref{['du_fig1']} with additional fits corresponding to adding each new data set one at a time.
• Figure 5: Same as Fig. \ref{['du_fig1']} with additional fits corresponding to adding each new data set two at a time.