J. Marshall

Software compensation in Particle Flow reconstruction

The particle flow approach to calorimetry benefits from highly granular calorimeters and sophisticated software algorithms in order to reconstruct and identify individual particles in complex event topologies. The high spatial granularity, together with analogue energy information, can be further exploited in software compensation. In this approach, the local energy density is used to discriminate electromagnetic and purely hadronic sub-showers within hadron showers in the detector to improve the energy resolution for single particles by correcting for the intrinsic non-compensation of the calorimeter system. This improvement in the single particle energy resolution also results in a better overall jet energy resolution by improving the energy measurement of identified neutral hadrons and improvements in the pattern recognition stage by a more accurate matching of calorimeter energies to tracker measurements. This paper describes the software compensation technique and its implementation in particle flow reconstruction with the Pandora Particle Flow Algorithm (PandoraPFA). The impact of software compensation on the choice of optimal transverse granularity for the analogue hadronic calorimeter option of the International Large Detector (ILD) concept is also discussed.

The Pandora multi-algorithm approach to automated pattern recognition in LAr TPC detectors

Pattern recognition is the identification of structures and regularities in data. In high energy physics, it is a vital stage in the reconstruction of events recorded by fine-granularity detectors. The development and operation of Liquid Argon Time Projection Chambers (LAr TPCs) for neutrino physics has created a need for new approaches to pattern recognition, in order to fully exploit the superb imaging capabilities offered by this technology. Whereas the human brain excels at identifying features in the recorded events, it is a significant challenge to develop an automated solution. The Pandora Software Development Kit (SDK) provides functionality to aid the process of designing, implementing and running pattern recognition algorithms. In particular, it promotes the use of a multi-algorithm approach to pattern recognition: individual algorithms each aim to address a specific task in a particular topology; a series of many tens of algorithms then carefully build-up a picture of the event and, together, provide a robust automated pattern recognition solution. Building on successful use of the Pandora SDK for pattern recognition at collider experiments, a sophisticated chain of algorithms has been created to perform pattern recognition for neutrino experiments utilising LAr TPCs like MicroBooNE. The input to the Pandora pattern recognition is a list of 2D Hits. The output from the chain of over 70 algorithms is a hierarchy of reconstructed 3D Particles, each with an identified particle type, vertex and direction. In this document, we present details of the Pandora pattern recognition algorithms used to reconstruct cosmic-ray and neutrino events in LAr TPCs. We also present metrics that assess the current reconstruction performance using simulated data from MicroBooNE.