<p> The high-energy universe is known to be violent. Ultra High Energy Cosmic Rays (UHECRs) have been observed with kinetic energies exceeding 10<sup> 20</sup> eV. Their origin, despite decades of observations, remains elusive. A unique probe of the sources and production mechanisms of these high energy cosmic rays can be neutrinos, since they are inevitably produced when high-energy protons interact. The IceCube Neutrino Observatory, located at the geographical South Pole in Antarctica, continuously monitors a total volume of 1 km<sup> 3</sup> of clear Antarctic ice for neutrino interactions. For this purpose, a total of 5160 optical sensors (photomultiplier tubes) have been melted deep into the glacier at depths between 1450m and 2450m. In 2013 IceCube reported one of its biggest discoveries, the observation of highly energetic neutrinos that are consistent with a possible extra-galactic origin. </p><p> In this dissertation we use IceCube data (recorded from 2012 to 2015) to study the spectral properties of this astrophysical neutrino flux with focus on electron and tau neutrino flavors. We developed a new neutrino identification and muon background rejection method using state-of-the-art machine-learning techniques, more specifically multi-class gradient boosted decision trees. In addition to enlarging the number of detected neutrino events (>10x increase over previous works), we lowered the energy threshold to below 1 TeV and thereby greatly improved upon the control and treatment of systematic uncertainties. The sample contains ~400 astrophysical electron and tau neutrinos, which increases the significance of the original discovery to beyond 8 standard deviations. We find the astrophysical neutrino flux to be well described by a single power-law consistent with expectations from Fermi-type acceleration of high-energy particles at astrophysical sources and obtain leading constraints on its properties. We further studied the possibility of additional spectral complexity, which significantly increases measurement uncertainties. No evidence for such scenarios was found. Finally we searched for a contribution from atmospheric neutrinos related to heavy meson (charm) decay in Earth's atmosphere and derive a flux upper limit of 4.8 times the benchmark pQCD flux prediction at 90% confidence level, dominated by systematic uncertainties, especially related to photon transport in the glacial ice.</p><p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10823307 |
| Date | 19 September 2018 |
| Creators | Niederhausen, Hans |
| Publisher | State University of New York at Stony Brook |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0019 seconds