Reconstruction of primary vertices in $pp$ collisions at energies of 900 GeV and 7 TeV with the ATLAS detector

TL;DR

This paper discusses the development and performance evaluation of primary vertex reconstruction algorithms used in the ATLAS detector for analyzing initial proton-proton collisions at 900 GeV and 7 TeV energies at the LHC.

Contribution

It presents new algorithms and their performance metrics for primary vertex reconstruction in early LHC collision data at different energies.

Findings

Reconstruction efficiencies are suitable for initial physics analyses.

Vertex position resolutions are quantified through pull distributions.

Algorithms perform reliably at different collision energies.

Abstract

The Large Hadron Collider (LHC) of the European Organisation for Nuclear Research (CERN) started its operation in Autumn of 2009. The initial run at a centre-of-mass energy of 900 GeV, has been followed by the on-going run at the energy of 7 TeV. While initially the probability of several proton-proton collisions to happen within the same bunch-crossing was approximately 10^-5, the level of the pile-up grows steadily with better focusing and squeezing of the LHC beams at collision point. Presented in this contribution is the performance of the primary vertex reconstruction algorithms used for analysis of the first collisions at the LHC. Different approaches used for the reconstruction of primary vertices in 900 GeV and 7 TeV collisions are presented. The efficiencies of the primary vertex reconstruction used for the first physics analyses of ATLAS are shown. The resolutions on positions of the reconstructed primary vertices are investigated by studying the distributions of pulls of distance between artificially created half-vertices. Implications of the ATLAS performancewith respect to primary vertex reconstruction for the on-going and future physics analyses are discussed.