Higgs Boson searches in gluon fusion and VBF using the H->WW decay mode

This page contains approved plots and results in the order as they appear in the CSC note. Only the CSC note contains all the relevant information and should thus be consulted if one of the plots is used.

Figure 1: Jet pT distributions (|η| < 2.5) for the H->WW signal sample with (red line) and without (gray filled area) pile-up. Left: Standard jets. Right: Track jets.

Figure 2: Left: The log Likelihood Ratio distributions for toy Monte Carlo corresponding to the H->WW->lνqq analysis at 10 fb^-1 for background-only outcomes (red) and signal-plus background outcomes with several values of the Higgs boson mass. Right: The same plot, but for the H + 2 j, H->WW->eνμν analysis.

Figure 3: Left: transverse opening angle Δφ_ll of the two leptons after preselection cuts. Middle : transverse momentum p_T^WW of the WW system after preselection cuts. Right: transverse mass M_T for events with Δφ_ll < 1.575 and p_T^WW > 20 GeV, in a fitted toy Monte Carlo outcome containing a Standard Model Higgs boson with M_H = 170 GeV, after 10 fb^-1 of integrated luminosity. The Likelihood Ratio in this outcome is 30.69, which is typical for this value of M_H and this luminosity.

Figure 4: Upper Left: The log Likelihood Ratio distributions for background-only toy Monte Carlo outcomes corresponding to 10 fb^-1 in the H + 0 j, H->WW->lνlν analysis. Upper Right: The corresponding pull distributions for M_H =170 GeV. Lower Left: The linearity of the mass determination. Lower Right: the expected significance for an integrated luminosity of 10 fb^-1

Figure 5: Pseudorapidity gap between tag jets (left top plot), invariant-mass distributions of tag jets (right top plot), azimuthal angle gap between tag jets (left bottom plot) and E_t of the third jet in VBF H->WW->μνμν Pythia events (m(H)=170 GeV). A requirement η1 η2<=0 is used in addition to the requirement jet E_t > 20 GeV.

Figure 6: The distribution of leading versus sub-leading jet weights in the events for signal (left plot) and the t t background (right plot). The plots are divided in three regions: (I) where there is a non-default b-tagging weight for more than one jet, (II) where there is only b-tagging information for one jet in the event and (III) where there are no jets with b-tagging information in the event.

Figure 7: An example fit to a toy Monte Carlo outcome corresponding to 10 fb^-1 of integrated luminosity in the H + 2 j, H->WW->eνμν channel. The pseudodata contains a Standard Model Higgs boson with a true mass of 170 GeV. Left: The Neural Network output distribution in the signal box, for events with 50 < M_T < 180 GeV. Right: The transverse mass distribution for events in the signal box with Neural Network output larger than 0.8. The Likelihood Ratio for this outcome was 22.62, which is typical if the signal model used as truth is derived from a Monte Carlo sample generated with SHERPA.

Figure 8: The Neural Network output distribution for the H + 2 j, H->WW->eνμν channel in the signal box. Right: The expected significance at 10 fb^-1 . The blue and magenta triangles were computed using a fixed extrapolation.

Figure 9: The Higgs boson transverse mass distribution for events in the defined physics signal region, corresponding to approximately 1 fb^-1 of data, as obtained from fully simulated Monte Carlo events. The solid (blue) curve is the total fit projection. The background contribution is represented by the dashed (red) curve. Right: The expected statistical sensitivity to Standard Model VBF H->WW->lνlν decays for 1 fb^-1 of ATLAS data, using the five-dimensional fit of Section 6.3.

Figure 10: Left: the reconstructed invariant mass of the M_lνqq system for fully simulated signal Monte Carlo events with a true Higgs boson mass of 300 GeV. The red dashed and black dotted curves show the mass peaks obtained when the jet energy scale is raised and lowered by 7% in the region with |η| < 2.5 and 15% elsewhere. The blue dot-dashed curve shows the result obtained when the jet energy resolution is smeared by 45%/sqrt(E) in the region with |η| < 2.5 and 63%/sqrt(E) elsewhere. Middle: A toy Monte Carlo outcome corresponding to 10 fb^-1 of integrated luminosity for the signal-like region of the H + 2 j, H->lνqq analysis described in Section 7. Here, the background from QCD multijets is assumed to be negligible. Right: the corresponding distribution in the control region.

Figure 11: The linearity of the mass determination for the combined fit of H + 0/2 j, H->WW->eνμν. Right: The expected significance at 10 fb^-1.