LO vs NLO comparisons for Z + jets: MC as a tool for background determination for NP searches at LHC

TL;DR

This paper compares LO and NLO simulation models for Z + jets production at the LHC, evaluating their accuracy in background estimation for new physics searches and analyzing the impact of different modeling choices.

Contribution

It provides a detailed comparison of LO and NLO models for Z + jets, focusing on the effects of matrix elements, MPI models, and parton shower tuning against Tevatron data.

Findings

LO and NLO models show different predictions for jet kinematics

Model tuning significantly affects background estimates

NLO models better match experimental data in certain regions

Abstract

The leptonic decays of the heavy gauge bosons W and/or Z provide a clear experimental signature at hadron colliders. The production of accompanying jets is an excellent signal to probe QCD, while also being the main background to many searches for new physics. Describing the complex final state of W or Z + jets is a theoretical challenge with most existing calculations combining matrix elements for high energy jet production with a parton shower for lower energy jet production. We focus on two models: SHERPA, which uses Leading Order matrix elements for boson and jet production; and POWHEG with HERWIG++, which uses a Next-To-Leading Order Matrix element for Z production. In order to isolate the impact of the matrix elements for jet production, it is first essential to constrain the differences in the rest of the calculation in each case: specifically, the Multiple Parton Interaction models, and the tuning of the parton shower interfaced to the matrix elements. We test all three aspects of these models against data from the Tevatron, and perform a study of some basic kinematic variables at the LHC energy.