Combined detection of supernova neutrino signals

TL;DR

This paper introduces a shape analysis method using a log likelihood ratio to improve supernova neutrino detection sensitivity, enabling better signal identification, combining data from multiple detectors, and extending prediction times for supernova events.

Contribution

The paper presents a novel shape analysis approach based on a log likelihood ratio for supernova neutrino detection, enhancing sensitivity and enabling detector data combination.

Findings

Shape analysis improves detection significance and prediction accuracy.

Combining data from multiple detectors increases sensitivity.

Method remains effective even with model mismatches.

Abstract

Supernova neutrino detection in neutrino and dark matter experiments is usually implemented as a real-time trigger system based on counting neutrino interactions within a moving time window. The sensitivity reach of such experiments can be improved by taking into account the time profile of the expected signal. We propose a shape analysis of the incoming experimental data based on a log likelihood ratio variable containing the assumed signal shape. This approach also allows a combination of potential supernova signals in different detectors for a further sensitivity boost. The method is tested on the NOvA detectors to study their combined sensitivity to the core-collapse supernova signal, and also on KamLAND, Borexino and SK-Gd as potential detectors of presupernova neutrinos. Using the shape analysis enhances the signal significance for supernova detection and prediction, as well as the sensitivity reach of the experiment. It also extends the supernova prediction time when applied to the presupernova neutrino signal detection. Enhancements achieved with the shape analysis persist even in the case when the actual signal doesn't match the expected signal model.