Presentation description

The IceCube Neutrino Observatory is an astroparticle detector located at the geographic south pole that was primarily designed to detect neutrinos, which are low mass particles that rarely interact with matter. These neutrinos have three flavors, which are named after their associated particles: muon, electron, and tau. Utilizing a cubic kilometer of ice, the observatory detects rare neutrino interactions with matter. When these neutrinos interact with the ice they emit a form of light radiation called Cherenkov radiation. The IceCube detectors, called DOMs (digital optical modules), can then detect as few as one photo of light, and record data from it. This data can then be utilized to reconstruct the neutrino's inelasticity, defined as the ratio of the neutrino's energy before interaction to the energy transferred to the ice. Our project specifically works with low energy muon neutrinos (1-300 Gev) which increases difficulty as particles with lower energy provide less data. Utilizing the DeepCore DOMs, which are a set of DOMs located in the bottom center of the detector with higher density, we are able to use convolutional neural networks (CNNs) to extract positional data for accurate event reconstruction. After reconstruction, this information can be used to analyze the ratio between neutrino and antineutrino occurrences, as inelasticity varies between the two. Another application is neutrino mass ordering analysis at low energy, which is determining the ordering for the three flavors of neutrinos. Utilizing this new inelasticity reconstruction method will be useful for these and other applications.