Results from the Pan-STARRS Search for Kilonovae: Contamination by Massive Stellar Outbursts

TL;DR

This study analyzes Pan-STARRS data to identify kilonovae, finds that massive stellar outbursts are major contaminants, and suggests photometric methods to distinguish them in gravitational wave follow-up searches.

Contribution

It provides the first comprehensive assessment of stellar outburst contamination in optical kilonova searches and proposes strategies for photometric identification.

Findings

175 transients within 200 Mpc with M > -16.5 identified

All kilonova candidates eliminated after follow-up

Massive stellar outbursts account for 55% of contaminants

Abstract

We present results from the Pan-STARRS optical search for kilonovae without the aid of gravitational wave and gamma-ray burst triggers. The search was conducted from 26 October 2019 to 15 December 2022. During this time, we reported 29,740 transients observed by Pan-STARRS to the IAU Transient Name Server. Of these, 175 were Pan-STARRS credited discoveries that had a host galaxy within 200 Mpc and had discovery absolute magnitudes M > -16.5. A subset of 11 transients was plausibly identified as kilonova candidates by our kilonova prediction algorithm. Through a combination of historical forced photometry, extensive follow-up, and aggregating observations from multiple sky surveys, we eliminated all as kilonova candidates. Rapidly evolving outbursts from massive stars (likely to be Luminous Blue Variable eruptions) accounted for 55% of the subset's contaminating sources. We estimate the rate of such eruptions using the ATLAS 100 Mpc volume-limited survey data. As these outbursts appear to be significant contaminants in kilonova searches, we estimate contaminating numbers when searching gravitational wave skymaps produced by the LIGO-Virgo-Kagra science collaboration during the Rubin era. The Legacy Survey of Space and time, reaching limiting magnitudes of m = 25, could detect 2-6 massive stellar outbursts per 500 deg^2 within a 4-day observing window, within the skymaps and volumes typical for binary neutron star mergers projected for Ligo-Virgo-Kagra Observing run 5. We conclude that while they may be a contaminant, they can be photometrically identified.