A QCD analysis of HERA and fixed target structure function data

TL;DR

Problem: precise determination of proton parton densities with uncertainties from HERA and fixed-target DIS data, including sea-quark asymmetry. Approach: a next-to-leading order QCD fit using DGLAP evolution in the MS̄ scheme with a massless variable-flavor framework, incorporating higher-twist corrections and nuclear effects; a comprehensive error analysis combines experimental, systematic, and scale uncertainties and provides a public data file. Findings: the fit describes the data well, yielding constrained gluon and quark momentum fractions, but the large-x d/u behavior remains poorly constrained; nuclear effects in neutrino scattering are consistent with charged-lepton data, and SM predictions for e^+ p NC/CC cross sections agree with ZEUS measurements.

Abstract

The parton momentum density distributions in the proton are determined from a next-to-leading order QCD analysis of structure functions measured at HERA and by fixed target experiments. Also included are data on the difference of the up and down anti-quark distributions. The uncertainties in the parton densities, structure functions and related cross sections are estimated from the experimental errors and those on the input parameters of the fit. Several QCD predictions obtained from this analysis are calculated and compared to data. The parton densites including the full error information is made available on a computer readable file.

Figures (15)

• Figure 1: The ratio of nuclear structure functions to those of a free nucleon versus $x$: (a) $F_2^d/F_2^N$ from ref:gomez; (b) $xF_3^{\nu Fe}/xF_3^{\nu N}$ from ref:eskola. The shaded bands correspond to the uncertainties on the ratios assumed in this analysis. Also indicated are the $x$ ranges of the $F_2$ and $xF_3$ data included in the QCD fit.
• Figure 2: The proton $F_2$ structure function versus $x$ at fixed values of $Q^2$ from ZEUS ref:zeusnv, H1 ref:h1qcd, E665 ref:e665, NMC ref:newnmc, BCDMS ref:bcdms and SLAC ref:slac. Only shown are the data included in the QCD analysis. The full curves show the result from the QCD fit. For clarity the constants given in brackets are added to $F_2$. The ZEUS, H1 and E665 data points are plotted with open and solid symbols for alternating $Q^2$ bins.
• Figure 3: The structure functions $F_2^p$ (left) and $F_2^d$ (right) from fixed target experiments ref:newnmcref:bcdmsref:slac versus $Q^2$ for $x > 0.1$, $Q^2 > 3$ and $W^2 > 7$ GeV$^2$. The full (dotted) curves correspond to the QCD fit results including (excluding) higher twist contributions. The $F_2^d$ curves include the correction for nuclear effects in deuterium described in the text. The $x$ values and multiplication factors (in brackets) given in the left-hand plot apply to both $F_2^p$ and $F_2^d$.
• Figure 4: The higher twist correction $H(x)$ as defined in Section \ref{['se:qcdfit']}. The full curve corresponds to the result of the QCD analysis. The quadratic sum of the error contributions (1) and (2) given Section \ref{['se:errors']} is drawn as the shaded band around the curve. The dashed curve shows $H(x)$ obtained from an alternative fit described in Section \ref{['se:checks']}. Also plotted are the results from an analysis of MRST ref:mrsht (asterisks). In the inset $H(x)$ from this analysis is compared to the result from the QCD fit of ref:marcv to the SLAC and BCDMS data.
• Figure 5: The $xF_3^{\nu Fe}$ structure function from CCFR ref:ccfrxf3 versus $Q^2$ at fixed values of $x > 0.1$. The full curve shows the QCD prediction corrected for nuclear effects on iron, described in the text. The shaded bands indicate the error on the QCD fit. For clarity $xF_3$ is multiplied by the factors indicated in brackets.