Supernovae with IceCube: direction and average neutrino energy determination

Abstract

Supernovae explosions are among the most powerful events known to occur in the universe. They are also to date the only known source of extrasolar neutrinos. Observing such an explosion in the neutrino sector would provide valuable information about the explosion mechanism of the star, as well as properties of the neutrino. The IceCube neutrino telescope monitors the Antarctic glacier for neutrino induced Cherenkov photons. Even though it was designed to detect high energy neutrinos, IceCube can detect large numbers of MeV neutrinos by observing a collective rise in all photomultiplier rates. This feature enables IceCube to detect outbursts of neutrinos from core collapse supernovae within the Milky Way. In case of a supernova in the centre of the galaxy, IceCube would be able to provide the highest statistics of all experiments world-wide, recording ≈40.000 times more neutrino events than recorded for the last observed supernova in 1987. The collective photomultiplier rate, however does not carry information about single neutrinos making it e.g. impossible to determine the energy and di- rection. Part of this thesis was dedicated to developing new methods to remedy this situation. In the course of this thesis, major contributions have been made to extend the functionality, increase the reliability and to improve the monitoring of the data acquisition system to detect core collapse supernovae. A newly introduced storage system of all recorded photons for an adjustable time in case of an alert opened new analysis opportunities. The passage of the neutrino wave front through the detector can in principle be monitored by triangulation even in the presence of a dark rate background, whenever the flux changes abruptly. This is, e.g., the case for large progeni- tor stars that end up in a black hole, shutting down the neutrino flux almost instantaneously. By using a proper likelihood description, a method has been developed that estimates the supernova direction with 20 degree uncertainty, if the effect of neutrino masses can be neglected and the flux ceases sufficiently fast at the time of black hole formation. The coincidence probability for observing Cherenkov light from O(10 cm) long positron tracks in the 17 m spaced light sensors lies only in the percent range. Nevertheless, given the large signal on top of the background, one can estimate the fraction of coincidences and thus determine the average neutrino energy.