Extracting the time of core-bounce from core-collapse supernova neutrino signals in current and next-generation neutrino detectors

Abstract

Core-collapse supernovae (CCSNe) are amongst the most rare and energetic events in the galaxy. In the Milky Way, they are predicted to happen as infrequently as 1.64±0.46 times per century. Over a duration of approximately ten seconds, a CCSN will convert ≈ 99% of its iron core’s gravitational binding energy into neutrinos. The initial wave of neutrinos is powered by the neutronization burst, which is generated by electron capture reactions on the collapsing core, which follows a critical time in the dynamics of a CCSN, core-bounce. It has been 36 years since a CCSN was observed via its neutrinos. With the observation of SN 1987A via its neutrino signal, a global effort has been undertaken to bring together all neutrino detectors under a common goal: providing an early alert to the astronomical community of an impending supernova and, if possible, triangulate to the CCSN using its neutrino signal. This effort is called the SuperNova Early Warning System (SNEWS). Triangulation simulations have recently seen tremendous success in determining where a supernova is positioned from its neutrinos, but these studies have

made use of high statistics detectors such as HyperK, JUNO, and DUNE. This work im- plements six analytic techniques into the detectors HALO and HALO-1kT, with the intent

of extracting a common reference time across all detectors to use for triangulation efforts. The common reference time chosen is the time of core-bounce (t0), as it is followed by a rapid rise in νe events within νe sensitive neutrino detectors. Our analysis made use of the

SNOwGLoBES event rate calculator, which quantifies event rates from supernova neu- trino signals, which was then simulated through each detector’s Monte Carlo simulation

code. Various supernova models were taken into consideration to account for systematic uncertainties between different mass progenitors, equations of state, etc. Our analysis determined that for HALO and HALO-1kT, a constant fraction discriminator (CFD) technique was optimal in extracting the time of core-bounce from the neutrino signal at close distances (< 3 kpc), while a negative log likelihood technique was optimal at further distances. At 1 kpc, HALO-1kT had a precision of 543 μs when using the CFD technique to extract t0, which falls within the precision required to triangulate effectively (< 1 ms, which HALO-1kT can obtain out to 3 kpc). With the intent of eventually implementing these techniques into all experiments involved in SNEWS 2.0, SNO+ was incorporated into our analysis in the later stages of this research. A preliminary exploration showed severely degraded performance in contrast to HALO-1kT, where the CFD technique could only obtain millisecond precision, not microsecond. Further analysis is encouraged.