Studying W+W- production at the Fermilab Tevatron with SHERPA

TL;DR

The paper validates SHERPA's merging of tree-level matrix elements with parton showers for $W^+W^-$ production at the Tevatron, using consistency tests across merging scales $Q_{\rm cut}$ and multiplicities $n_{\max}$, and comparing to fixed-order NLO results from MCFM and to other event generators. It finds that inclusive observables are largely stable under these variations and shows general agreement with NLO within about 20%, while the parton shower component introduces additional QCD radiation in jet-rich regions. The study also highlights the significant impact of spin correlations in $W$ decays on leptonic observables and demonstrates that higher-multiplicity matrix elements improve high-$p_T$ regions. Overall, SHERPA provides a robust ME+PS framework that competes with NLO calculations and offers insight into QCD radiation patterns relevant for Tevatron analyses.

Abstract

The merging procedure of tree-level matrix elements with the subsequent parton shower as implemented in SHERPA will be studied for the example of W boson pair production at the Fermilab Tevatron. Comparisons with fixed order calculations at leading and next-to-leading order in the strong coupling constant and with other Monte Carlo simulations validate once more the impact and the quality of the merging algorithm and its implementation.

Figures (27)

• Figure 1: The $p_T$ distribution of the $W^+$ boson and its dependence on $Q_{\rm cut}$, chosen to be $15$, $30$ and $80$ GeV (from top to bottom). The black solid line shows the SHERPA prediction obtained with $n_{\rm max}=2$, the black dashed one is the reference obtained as the mean of different $Q_{\rm cut}$ runs and the coloured lines indicate the different multiplicity contributions. The lower part of the plots exhibits the normalized difference of the prediction with respect to the reference. Cuts and input parameters are specified in the appendices.
• Figure 2: The $p_T$ distribution of the $W^+W^-$ system under merging scale variation. The cut indicated through a vertical dashed-dotted line has been chosen as $Q_{\rm cut}=15$, $30$ and $80$ GeV (from top to bottom). The black solid line shows the SHERPA prediction obtained with $n_{\rm max}=2$, the black dashed one is the reference obtained as the mean of different $Q_{\rm cut}$ runs and the coloured lines indicate the different multiplicity contributions. The lower part of the plots exhibits the normalized difference of the prediction with respect to the reference. Cuts and input parameters are specified in the appendices.
• Figure 3: Differential $1\to0$ jet rate $Q_1$, $2\to1$ jet rate $Q_2$ and $3\to2$ jet rate $Q_3$ (left to right) for the SHERPA$n_{\rm max}=2$ configuration. The cut has been chosen to be $15$, $30$ and $80$ GeV (from top to bottom). The black solid line shows the total result, the black dashed one is the reference obtained as the mean of different $Q_{\rm cut}$ runs and the coloured lines indicate the different multiplicity contributions. The vertical dashed dotted line indicates the separation cut position. The lower part in all plots pictures the normalized difference of the corresponding prediction with respect to the reference. For input parameters and cuts, see Apps. \ref{['app_input']} and \ref{['app_cuts']}.
• Figure 4: The $p_T$ distribution of the $W^+$ boson in dependence on the variation of the maximal jet number. The comparison is to a (black dashed) reference curve obtained with $n^{\rm ref}_{\rm max}=n_{\rm max}-1$. The cut has been chosen to be $15$ GeV. In both plots the black solid line shows the total result obtained with SHERPA. The coloured lines indicate the different multiplicity contributions. The lower part in both plots visualizes the normalized difference of the corresponding prediction with respect to the reference. For input parameters and cuts, see Apps. \ref{['app_input']} and \ref{['app_cuts']}.
• Figure 5: The $H_T$ distribution and its dependence on the variation of the maximal jet number. The separation cut has been chosen to be $15$ GeV. The green solid line shows the SHERPA prediction obtained with $n_{\rm max}=1$, the lighter dashed and the grey dotted one stand for the $n_{\rm max}=2$ and $n_{\rm max}=3$ prediction, respectively; the darkgreen dashed-dotted curve pictures the pure shower performance of SHERPA starting off with the lowest order matrix element. The lower part of the plot shows the normalized differences with respect to the $n_{\rm max}=1$ case. For input parameters and cuts, see Apps. \ref{['app_input']} and \ref{['app_cuts']}.