Sabrina Sacerdoti is a Ph.D. student from Universidad de Buenos Aires, on a three-month MCnet studentship in the Durham team.

At the Large Hadron Collider (LHC), each high-energy collision can contain sprays of hadrons, collected into objects called jets. Events with high jet-multiplicity present one of the most significant and most notoriously difficult to model processes at the LHC. These backgrounds were already important in searches for beyond the Standard Model physics in the first run of the LHC, and they will be even more important in the second run, where the higher center of mass energy will lead to higher jet multiplicities in the final states. Understanding these processes with precision is also necessary for Standard Model measurements of high-jet-multiplicity final states, including measurements that are critical for the understanding of the newly discovered Higgs boson (e.g. the associated production of top quarks and additional heavy flavor quarks). The old workhorse generators, Pythia and Herwig, provide only a two-to-two, leading order (LO) description of the hard scatter, and they only include the possibility of a leading-log (LL) approximation in the parton shower. Some newer and more complicated, but CPU-challenging generators, including Sherpa, Alpgen, and MadGraph, allow the use of many-parton final states modeled at leading order. Recently, some Monte Carlo generator advances have allowed the possibility of using next-to-leading order (NLO) matrix elements and extensions to next-to-leading log (NLL) parton showers are being actively pursued, making explicit sensitive tests of such aspects a relevant topic for experimental studies. Beyond simply jet counting, an open question remains which observables are most sensitive to the differences between LO and NLO matrix elements, between differences in matching schemes (particularly including the new schemes available in Pythia8), and potentially between LL and NLL parton showers. The primary aim of this project is to construct observables, first in four-jet final states and then in many-jet final states, which are particularly sensitive to these differences. This naturally includes understanding how to produce Monte Carlo events with a reasonable distribution throughout the phase space, since the production of inclusive multi-jet events frequently fails to populate the interesting regions of kinematic phase space with a sufficiently high rate. Another open question remains whether the gain produced by higher multiplicity matrix elements is no longer detectable after a certain point – that is, whether two-to-six matrix element production is detectably different from two-to-ten matrix element production, after the effects of a parton shower have been included. This project should include an examination of these issues, which will provide critical feedback to the LHC experiments and to theorists hoping to model these backgrounds. As a further step of this project, it is important to test the generality of these conclusions in events with bosons in the final state. The addition of a Z, W or top (pair) introduces a second scale in the event, which can make the inclusion of these higher order effects more important. Nonetheless, if it is unnecessary to include very high multiplicity matrix elements, then enormous computing resources can be saved.

Sabrina Sacerdoti
March, 2014 to June, 2014