Preview

PDF, English - main document Download (52MB) | Terms of use

Abstract

The particle flow reconstruction concept is based on a set of pattern recognition algorithms promising to deliver unprecedented jet energy resolution in a future lepton collider experiment. One of the key requirements for this concept is highly granular calorimetry, capable of revealing the sub-structure of particle showers. The CALICE collaboration has developed the highly granular Analog Hadron Calorimeter (AHCAL) prototype, a steel sampling calorimeter featuring ~22000 readout channels of scintillating tiles coupled to silicon photomultipliers (SiPMs). During extensive beam test campaigns at the SPS CERN in 2018, the prototype has been successfully operated in muon, electron and pion beams proving feasibility of the technology and scalability to a collider detector. The first part of this thesis focuses on the characterisation and calibration of the AHCAL prototype. For all channels excellent signal-to-noise ratios, very good uncalibrated response uniformities and stable operation over time and for different operating modes are demonstrated. In the second part, the Pandora particle flow algorithm (PandoraPFA) framework is applied to AHCAL prototype data and Monte Carlo simulations. On the basis of extensive studies with regard to the limiting effects of particle flow reconstruction in single and two hadron events, the reliability of performance projections for future lepton collider experiments has been further validated with realistic detector data and detailed simulations. In addition, profound understanding of the PandoraPFA sub-algorithm interplay and the impact of counteracting increased noise levels on the particle flow reconstruction performance has been gained by studying modified PandoraPFA settings and increased energy thresholds on calorimeter channel level for AHCAL prototype data and jet simulations in a potential future lepton collider experiment.