A Jet Based Approach to Measuring Soft Contributions to Proton-Proton Collisions with the CMS Experiment

The early running stages of the Large Hadron Collider (LHC) offered perfect conditions to study the Underlying Event (UE) in proton collisions. This term denotes all effects in collider events that are not directly related to the hard partonic interaction. As these effects are present in all collisions, it is necessary to include them in Monte Carlo simulations. It is however not possible to calculate UE contributions with means of perturbation theory, which makes the application of phenomenological models necessary. These models have to be tuned to data, a process that has to be repeated at every newly accessed collision energy. At the start of the LHC program, different tunes have been available that were derived from Tevatron data and extrapolated to LHC energies, yielding large deviations in the predictions of UE contributions, depending on the scaling behaviour of the models. The influence of different tunes for the underlying event is demonstrated by investigating their influence on the inclusive jet spectrum. This quantity is a basic observable at hadron colliders yet the theoretical prediction in Monte Carlo generators is available to the particle level only in leading order of perturbative QCD. Therefore, correction factors have to be applied to next-to-leading matrix-element calculations to account for effects of the UE and hadronization. It is shown in this thesis that these correction factors strongly depend on the applied UE model and that by using multiple models, a systematic uncertainty on the correction factors can be estimated. Furthermore, a new and complimentary method to investigate the underlying event is applied for the first time in this work. While traditionally, these effects have been studied by investigating the geometrical area transverse to a leading object in the event, the new approach uses the ratio of the transverse momenta of all jets in the event divided by their areas. This is the first exertion of the concept of jet areas in this context. In order to cover the phase space to particle momenta as low as possible, reconstructed tracks are used as input for the analysis. The silicon tracking detector of the CMS experiment proves to be an outstanding tool for this kind of measurement. Data taken at two different center-of-mass energies are studied: 0.9 TeV and 7 TeV recorded in 2009 and 2010 respectively. It is shown that all Monte Carlo tunes that were produced prior to LHC operation underestimate the event activity, a result that is in line with other analyses. A new model that includes first LHC results is tested as well, showing promising results. The new method of quantifying soft activity in hadron collisions applied here for the first time offers a new perspective also towards the subtraction of hadronic noise from pile-up events, a task that will benefit from the experience gained in this work.