NLO Uncertainties in Higgs + 2 jets from Gluon Fusion

TL;DR

This work provides a rigorous assessment of perturbative uncertainties for Higgs production with two jets via gluon fusion in the presence of VBF-like cuts. By applying the ST method to exclusive 2-jet bins, the authors quantify yield and migration uncertainties arising from jet-bin boundaries and Sudakov logarithms, and compare ST with the efficiency method. Their results show that tightening jet-veto cuts can dramatically increase theoretical uncertainties in the ggF background, potentially offsetting gains in ggF–VBF separation. The study offers practical guidance for incorporating these uncertainties into experimental strategies for isolating VBF Higgs production and lays groundwork for future resummation refinements.

Abstract

A central ingredient in establishing the properties of the newly discovered Higgs-like boson is to isolate its production via vector boson fusion (VBF). With the typical experimental selection cuts, the VBF sample is contaminated by a 25% fraction from Higgs + 2 jet production via gluon fusion (ggF) which has large perturbative uncertainties. We perform a detailed study of the perturbative uncertainties in the NLO predictions for pp -> H+2 jets via ggF used by the ATLAS and CMS collaborations, with the VBF selection cuts of their current H -> gamma gamma analyses. We discuss in detail the application of the so-called "ST method" for estimating fixed-order perturbative uncertainties in this case, and also consider generalizations of it. Qualitatively, our results apply equally to other decay channels with similar VBF selection cuts. Typical VBF selections include indirect restrictions or explicit vetoes on additional jet activity, primarily to reduce non-Higgs backgrounds. We find that such restrictions have to be chosen carefully and are not necessarily beneficial for the purpose of distinguishing between the VBF and ggF production modes, since a modest reduction in the relative ggF contamination can be easily overwhelmed by its quickly increasing perturbative uncertainties.

Figures (11)

• Figure 1: Exclusive 2-jet cross section using the ATLAS VBF selection for various scale choices as a function of $p_{THjj}^\mathrm{cut}$ (left panel) and $\pi-\Delta\phi_{H\!-\!jj}^\mathrm{cut}$ (right panel).
• Figure 2: Inclusive 2-jet cross section over a range of $\mu_r/m_H$ for ATLAS VBF selection (left panel) and CMS loose selection (right panel). The three curves show different values of $\mu_f$. The blue solid, green dotted, and green dashed curves correspond to $\mu_f=m_H$, $\mu_f=2m_H$, and $\mu_f=m_H/2$, respectively. The uncertainty bars show the inclusive 2-jet scale variation uncertainty.
• Figure 3: Inclusive 3-jet cross section as a function of $p_{THjj}^\mathrm{cut}$ (left panel) and $\pi -\Delta\phi_{H\!-\!jj}^\mathrm{cut}$ (right panel) for the ATLAS VBF selection. The outer solid green lines show the inclusive 3-jet scale variation uncertainty after symmetrization.
• Figure 4: Perturbative uncertainties in the exclusive 2-jet cross section with the ATLAS VBF selection as a function of $p_{THjj}^\mathrm{cut}$ (left panel) and $\pi-\Delta\phi_{H\!-\!jj}^\mathrm{cut}$ (right panel) for different choices of the correlation parameter $\rho$. Our default choice is $\rho = 0$.
• Figure 5: Exclusive 2-jet efficiency for different schemes in the efficiency method for $p_{THjj}^\mathrm{cut}$ (left panel) and $\pi-\Delta\phi_{H\!-\!jj}^\mathrm{cut}$ (right panel) using the ATLAS VBF selection.