Time-Dependent Point Source Search Methods in High Energy Neutrino Astronomy

TL;DR

This paper develops maximum-likelihood methods for detecting time-dependent neutrino fluxes from point sources, improving sensitivity for short-duration bursts in high energy neutrino astronomy.

Contribution

It introduces new methods for searching for both known and unknown time-dependent neutrino signals, enhancing detection efficiency over traditional time-integrated approaches.

Findings

Short-duration bursts require up to ten times fewer events for detection.

Methods are effective for both known and unknown burst time dependence.

Simulations demonstrate improved sensitivity in neutrino detection.

Abstract

We present maximum-likelihood search methods for time-dependent fluxes from point sources, such as flares or periodic emissions. We describe a method for the case when the time dependence of the flux can be assumed a priori from other observations, and we additionally describe a method to search for bursts with an unknown time dependence. In the context of high energy neutrino astronomy, we simulate one year of data from a cubic-kilometer scale neutrino detector and characterize these methods and equivalent binned methods with respect to the duration of neutrino emission. Compared to standard time-integrated searches, we find that up to an order of magnitude fewer events are needed to discover bursts with short durations, even when the burst time and duration are not known a priori.