A significant number of measurements and searches at the Large Hadron Collider (LHC) include events with production of many hadrons, which are grouped into objects called jets. While some modern generators, including MadGraph, Sherpa, and Alpgen, are able to model the production of additional jets with a many-parton final state in the matrix element, all rely on the use of a parton shower to model the production of additional softer jets and the sub-structure of all jets in the event. In order to test the shower development, one set of variables that has been proposed includes a re-clustering of the jets in the event, using a kT jet algorithm or similar method, until a di-jet final state is recovered. The clustering history of the event can then be used to better understand the topology of the final state. Such observables have not yet been used at the LHC for any measurement, though similar constructions (e.g. through the ‘razor’ variables) have been used in searches for new physics by both ATLAS and CMS. One aspect of this project is to develop these observables, which attempt to project out coherent emission off of multi-parton systems and identify kinematics for which one-to-three (or more) branching ought to be important. This can also lead to the development of observables or selections based on the clustering history that are particularly sensitive to differences in the modeling of matrix elements (e.g. LO vs NLO two-to-two and LO vs NLO two-to-N) or in the parton shower. Whether these or other observables can be made particularly useful by the inclusion of ratios between energies (e.g. 2.76 TeV vs 7 TeV vs 8 TeV) will also be explored. In developing these observables, it will be important to study the potential gain, or harm, from clustering already reconstructed and calibrated jets rather than all the individual particles in the event. Similar ideas have already been used for the construction of fat jets from narrow jets by running an additional jet algorithm with the calibrated jets as inputs. In this context, the jet threshold for inclusion in re-clustering should be studied, and whether observables are sensitive to that threshold. There is a key aspect of the shower development that this project aims to further study. The parton splitting in the shower is driven largely by the strength of the strong force coupling, αS. A frequently raised question is what the order of the coupling used in the shower should be, and whether that coupling need be the same order as that of the matrix element. The standard view is that the same order and scale should be used for both the parton scale and matrix element, and that the order of αS should be determined by the order of the generator. However, it has become clear in the last few years that the use of higher-order αS running can provide better modeling of observables at the LHC. This project should study whether using different orders for αS in the matrix element and parton shower can lead to inconsistencies that are actually observable in the final state. One other aspect of the parton shower that needs to be better understood is the production of heavy flavor (bottom, charm, or indeed top) quarks via off-shell gluon splitting, and whether further modifications of the parton shower might be necessary. As jet energies during run two will be commonly far higher than the masses of top quarks and bosons, these particles can be produced during a parton shower and lead to interesting features in the structure of jets. Whether these effects are important can be explored as an additional feature of this project. Supervisor: P. Skands
Emma Kuwertz, CERN, four months from May 2014.
May, 2014 to September, 2014