Master thesis defence by Anders Hammer Holm

Title: Starting Run2 at CERNs LHC - – Statistical approaches to electron identification and detector alignment in ATLAS

This thesis describes two analyses carried out in preparation for, and in the earliest days of, LHC Run 2. The analyses are not directly related, but both deal with issues affecting the ATLAS Inner Detector.

A novel method for detecting electrons with the Transition Radiation Tracker (TRT) is presented and tested. This method is motivated by the challenges presented by the high luminosity and dense bunch spacing of LHC Run 2 as well as the fact that the gas in some of the TRT straws has been changed following Run 1. These challenges require more dedicated electron identification, through a likelihood method that requires input parameters measured in data. Z decay events is a viable source of electrons, but a sufficient amount of this data is not available early in Run 2. The method presented is able to estimate the probabilities of HT hits for electrons and non-electrons in a given sample, as well as the fraction of electrons.

A new method of measuring pseudorapidity dependant misalignments, caused by systematic deformations (so-called weak modes) in the ATLAS inner detector is presented. This method is based on based on fitting the mass peaks of the decay of K-short particles to charged pions. Z decay is often used in alignment procedures, but the use of K-short represents a way of performing alignment earlier in the run, since sufficient K-short data becomes available on the first day.