A Precursor Plateau and Pre-Maximum [O II] Emission in the Superluminous SN2019szu: A Pulsational Pair-Instability Candidate

TL;DR

SN2019szu, a superluminous supernova, exhibited a unique pre-explosion plateau and early forbidden [O II] emission, likely caused by pulsational pair-instability ejections, providing insights into pre-supernova mass loss.

Contribution

This study presents the first detailed analysis of early forbidden emission lines and a pre-explosion plateau in a superluminous supernova, supporting a pulsational pair-instability origin.

Findings

Detection of early forbidden [O II] emission during the rising phase.

Identification of a pre-explosion plateau lasting about 40 days.

Evidence for circumstellar material ejected months before explosion.

Abstract

We present a detailed study on SN2019szu, a Type I superluminous supernova at $z=0.213$, that displayed unique photometric and spectroscopic properties. Pan-STARRS and ZTF forced photometry shows a pre-explosion plateau lasting $\sim$ 40 days. Unlike other SLSNe that show decreasing photospheric temperatures with time, the optical colours show an apparent temperature increase from $\sim$15000 K to $\sim$20000 K over the first 70 days, likely caused by an additional pseudo-continuum in the spectrum. Remarkably, the spectrum displays a forbidden emission line even during the rising phase of the light curve, inconsistent with an apparently compact photosphere. We show that this early feature is [O II] $\lambda\lambda$7320,7330. We also see evidence for [O III] $\lambda\lambda$4959, 5007, and [O III] $\lambda$4363 further strengthening this line identification. Comparing with models for nebular emission, we find that the oxygen line fluxes and ratios can be reproduced with $\sim$0.25 M$_{\odot}$ of oxygen rich material with a density of $\sim10^{-15} \rm{g cm}^{-3}$. The low density suggests a circumstellar origin, but the early onset of the emission lines requires that this material was ejected within the final months before the terminal explosion, consistent with the timing of the precursor plateau. Interaction with denser material closer to the explosion likely produced the pseudo-continuum bluewards of $\sim$5500 \AA. We suggest that this event is one of the best candidates to date for a pulsational pair-instability ejection, with early pulses providing the low density material needed for the forbidden emission line, and collisions between the final shells of ejected material producing the pre-explosion plateau.