Extracting Higgs boson couplings from LHC data

TL;DR

This paper demonstrates that by combining multiple Higgs production and decay channels at the LHC with mild theoretical bounds on HVV couplings, one can extract absolute Higgs couplings and the total width, not just ratios. It develops a fit framework that includes possible new loop contributions and undetected decays, quantifies systematic uncertainties, and shows that 10–40% precision on couplings is achievable for intermediate-mass Higgs with 300 fb^-1. It also explores MSSM scenarios, showing that significant deviations from SM predictions can be detected, especially via weak-boson-fusion channels, and quantifies the regions of MSSM parameter space where the SM can be distinguished. The results highlight the LHC's potential to illuminate Higgs-sector structure and guide future collider studies.

Abstract

We show how LHC Higgs boson production and decay data can be used to extract gauge and fermion couplings of Higgs bosons. We show that very mild theoretical assumptions, which are valid in general multi-Higgs doublet models, are sufficient to allow the extraction of absolute values for the couplings rather than just ratios of the couplings. For Higgs masses below 200 GeV we find accuracies of 10-40% for the Higgs boson couplings and total width after several years of LHC running. Slightly stronger assumptions on the Higgs gauge couplings even lead to a determination of couplings to fermions at the level of 10-20%. We also study the sensitivity to deviations from SM predictions in several supersymmetric benchmark scenarios as a subset of the analysis.

Figures (6)

• Figure 1: Relative precisions of fitted new partial widths as a function of the Higgs boson mass assuming SM rates and 30 fb$^{-1}$ at each of two experiments (left) and 300 fb$^{-1}$ at each of two experiments for all channels except WBF, for which 100 fb$^{-1}$ is assumed (right). The new partial width can be due to new particles in the loops for $H\to\gamma\gamma$ and $gg\to H$ or due to unobservable decay modes. See text for details. Here we make the weak assumption that $g^2(H,V)<1.05 \cdot g^2(H,V,{\rm SM})$ ($V=W,Z$).
• Figure 2: Relative precision of fitted Higgs couplings-squared as a function of the Higgs boson mass for the 2$\times$30 fb$^{-1}$ (left) and the 2$\times$300 + $2\times$100 fb$^{-1}$ (right) luminosity scenarios for SM rates. Here we make the weak assumption that $g^2(H,V)<1.05 \cdot g^2(H,V,{\rm SM})$ ($V=W,Z$) but allow for new particles in the loops for $H\to\gamma\gamma$ and $gg\to H$ and for unobservable decay modes. See text for details.
• Figure 3: As in Fig. \ref{['fig:fit']}, but with more restrictive assumptions. Here we assume that $g^2(H,W)=g^2(H,W,{\rm SM})\pm 5\%$ and $g^2(H,W)/g^2(H,Z)=g^2(H,W,{\rm SM})/g^2(H,Z,{\rm SM})\pm 1\%$. We also assume that no new particles run in the loops for $H\to\gamma\gamma$ and $gg\to H$, so that these couplings are fixed in terms of the couplings of the SM particles in the loops. As in Fig. \ref{['fig:fit']}, additional decays of the Higgs boson are fitted with a partial width for undetected decays (not shown).
• Figure 4: Fit within the MSSM $m_h^{\rm max}$ scenario in the $M_A$--$\tan\beta$ plane for three luminosity scenarios. The two panels show the region (to the left of the curves) which would yield a $\geq 5\sigma$ ($\Delta\chi^2\geq 25$) or $\geq 3\sigma$ ($\Delta\chi^2\geq 9$) discrepancy from the SM. The mostly-horizontal dotted lines are contours of $m_h$ in steps of 5 GeV.
• Figure 5: Fit within the MSSM $m_h^{\rm max}$ scenario in the $M_A$--$\tan\beta$ plane for the 2$\times$300 fb$^{-1}$ luminosity scenario. The $5\sigma$ lines are shown using WBF channels only (long--dashed), all channels except WBF (short--dashed), and the full fit (solid line).