Multiple Parton Interactions Studies at CMS

TL;DR

CMS presents a comprehensive study of Multiple Parton Interactions at the LHC up to 7 TeV, focusing on underlying event dynamics and forward energy flow, and outlines strategies for directly measuring double parton scattering in high-pT topologies. The work compares traditional and novel UE observables with a suite of PYTHIA tunes (and PHOJET in forward analyses), using Bayesian unfolding to correct detector effects and exploring PDF dependencies to constrain MPI modelling. It reveals rapid growth of underlying activity with the event scale and energy, a central saturation behavior at high pT, and forward-region data that provide complementary constraints for global tunes. Additionally, the paper discusses the DPS program in high-pT regimes, emphasizing observables that discriminate MPI from single-parton processes and highlighting the 3jet+gamma channel as a viable probe for future measurements.

Abstract

This paper summarizes the Multiple Parton Interactions studies in CMS, focusing on the already performed low pT QCD measurements up to highest centre of mass energies of 7 TeV and discussing the plans for the direct measurement of the multiple high-pT scatterings. The underlying event in pp interactions is studied measuring the charged multiplicity density and the charged energy density in the transverse region, which is defined considering the azimuthal distance of the reconstructed tracks with respect to the leading track-jet of the event, defined from tracks according to a jet clustering algorithms. In addition, we present the measurement of the underlying event using the jet-area/median approach, demonstrating its sensitivity to different underlying event scenarios. Observations in the central region are complemented by the mea- surement of the energy flow in the forward direction for minimum bias and central di-jet events. We compare our underlying event and forward results with the predictions from different Monte Carlo event generators and tunes, aiming to best parametrize the multiple parton interaction energy de- pendence starting from the Monte Carlo tunes developed to best fit the charged particle spectra measured at central rapidities. Finally we discuss the strategy to directly measure the multiple particle interactions rate focusing on the topologies with two hard scatterings in the same event.

Figures (4)

• Figure 1: (Upper plots) average multiplicity and average scalar $\sum p_T$ in the transverse region as a function of the leading track-jet $p_T$, for data at $\sqrt{s} = 0.9$ TeV and $\sqrt{s} =7$ TeV. (Bottom left plot) normalized scalar $\sum p_T$ distribution in the transverse region for data at $\sqrt{s} = 0.9$ TeV and $\sqrt{s} =7$ TeV; the leading track-jet is required to have $p_T > 3$ GeV/c. Predictions from PYTHIA-6 tune Z1 and PYTHIA-8.135 Tune 4C are compared to the corrected data. The inner error bars indicate the statistical uncertainties affecting the measurements, the outer error bars thus represent the statistical uncertainties on the measurements and the systematic uncertainty affecting the MC predictions added in quadrature. (Bottom right plot) normalized median of $p_T$ over jet area for track-jets reconstructed from collision data at $\sqrt{s} = 0.9$ TeV (black circles). Predictions from several PYTHIA-6 tunes and PYTHIA-8 Tune 1 are compared to data.
• Figure 2: Energy flow in the minimum bias (left) and di-jet (right) samples as a function of $\eta_{obs}$ at $\sqrt{s}=7$ TeV. Uncorrected data are shown as points, the histograms correspond to the MC predictions. Error bars corresponds to statistical errors. The shaded bands in these plots represent the systematic uncertainties of the measurements, which are largely correlated point-to-point.
• Figure 3: Definition of azimuthal angle between pairs, together with typical configurations of double-bremsstrahlung (left) and double-parton scattering events (right).
• Figure 4: Differential cross section shape as a function of $\Delta \phi^{(-)}$ (upper plots) and $\Delta \phi^{(+)}$ (bottom plots) variables. Predictions from PYTHIA 8.108 (Default scenario) and HERWIG 2.2.0 (left panel) and from three different Pythia settings (right panel) shown.