A DECam Search for Explosive Optical Transients Associated with IceCube Neutrinos

TL;DR

This study used DECam to follow up on IceCube neutrino alerts to find optical counterparts, specifically core-collapse supernovae, and assessed the likelihood of association based on observational data and statistical analysis.

Contribution

It demonstrates the feasibility of using DECam for optical follow-up of neutrino alerts and provides constraints on the association rate between CC SNe and high-energy neutrinos.

Findings

Two candidate CC SNe coincident with neutrino alerts were found.

No statistically significant excess of CC SNe was detected in the neutrino regions.

Expected detection of an excess at 3σ confidence level requires 60-200 alerts.

Abstract

In this work, we investigate the likelihood of association between realtime, TeV-PeV energy neutrino alerts from IceCube and optical counterparts in the form of core-collapse supernovae (CC SNe). The optical follow-up of IceCube alerts requires two main instrumental capabilities: (1) deep imaging, since 73\% of neutrinos would come from CC SNe at redshifts $z > 0.3$, and (2) a large field of view (FoV), since typical IceCube muon neutrino pointing accuracy is on the order of $\sim1$~deg. With Blanco/DECam ($gri$ to 24th magnitude and $2.2$~deg diameter FoV), we performed a triggered optical follow-up observation of two IceCube alerts, IC170922A and IC171106A on $\sim6$~nights during the $\sim3$~weeks following each alert. For the IC170922A (IC171106A) follow-up observations, we expect that 12.1\% (9.5\%) of coincident CC SNe at $z \lesssim 0.3$ are detectable, and that on average, 0.23 (0.07) unassociated SNe in the neutrino 90\% containment regions also pass our selection criteria. We find two candidate CC SNe that are temporally coincident with the neutrino alerts in the FoV, but none in the 90\% containment regions, which is statistically consistent with expected rates of background CC SNe for these observations. If CC SNe are the dominant source of TeV-PeV neutrinos, we would expect an excess of coincident CC SNe to be detectable at the $3\sigma$ confidence level using DECam observations similar to those of this work for $\sim60$ ($\sim200$) neutrino alerts with (without) redshift information for all candidates.