MRST2001: partons and alpha_S from precise deep inelastic scattering and Tevatron jet data

TL;DR

This study performs a comprehensive NLO global analysis to determine proton parton distributions and the strong coupling alpha_S(M_Z^2) by incorporating the latest high-precision DIS data from HERA and Tevatron jet data, along with updated neutrino, Drell-Yan, and deuterium measurements. Using a global DGLAP fit with TR variable-flavor scheme and equal data weighting, the authors find alpha_S(M_Z^2)=0.119 with experimental and theoretical uncertainties of ±0.002 and ±0.003, respectively, and they impose a refined, extended gluon parameterization that allows a negative small-x gluon at the input scale to accommodate the new data. The Tevatron jets provide strong constraints on the high-x gluon, while the small-x region remains challenging, leading to a nuanced picture where jet and DIS data together stabilize the fit and reduce gluon uncertainties. The results yield the most precise NLO parton set to date and offer a robust baseline for collider predictions, while also highlighting potential improvements from future NNLO analyses.

Abstract

We use all the available new precise data for deep inelastic and related hard scattering processes to perform NLO global parton analyses. These new data allow an improved determination of partons and, in particular, the inclusion of the recent measurements of the structure functions at HERA and of the inclusive jets at the Tevatron help to determine the gluon distribution and alpha_S better than ever before. We find a somewhat smaller gluon at low x than previous determinations and that alpha_S(M_Z^2) = 0.119 +/- 0.002 (expt.) +/- 0.003 (theory).

Figures (16)

• Figure 1: MRST2001 partons at $Q^2=10~{\rm GeV}^2$ and $Q^2 = 10^4~{\rm GeV}^2$.
• Figure 2: Comparison of the MRST2001 partons with those of MRST99$(g\!\uparrow)$MRST99 at $Q^2=10~{\rm GeV}^2$ and $Q^2=10^4~{\rm GeV}^2$.
• Figure 3: Comparison of the MRST2001 prediction of $F_2(x,Q^2)$ with data and with MRST99 for $x= 0.00005-0.00032$. The error bars show statistical and systematic errors added in quadrature.
• Figure 4: Comparison of the MRST2001 prediction of $F_2(x,Q^2)$ with data and with of MRST99 for $x= 0.0004-0.0025$. The data points are as indicated in Fig. 3.
• Figure 5: Comparison of the MRST2001 prediction of $F_2(x,Q^2)$ with data and with MRST99 for $x= 0.0032-0.0175$. The data points are as indicated in Fig. 3.