Yellow supergiants as supernova progenitors: an indication of strong mass loss for red supergiants?

TL;DR

This study investigates how increased mass-loss rates in red supergiants can explain the observed positions of yellow supergiant supernova progenitors, reconciling theory with observations.

Contribution

It demonstrates that higher mass-loss rates during the RSG phase enable single star models to match observed supernova progenitor positions in the HR diagram.

Findings

Higher mass-loss rates lead to models ending as yellow supergiants.

Models with increased mass loss match observed supernova progenitor positions.

Enhanced mass loss can explain the diversity of supernova progenitors.

Abstract

The increasing observed number of supernova events allows for finding ever more frequently the progenitor star in archive images. In a few cases, the progenitor star is a yellow supergiant star. The estimated position in the Hertzsprung-Russell diagram of these stars is not compatible with the theoretical tracks of classical single star models. According to several authors, the mass-loss rates during the red supergiant phase could be underestimated. We study the impact of an increase of these mass-loss rates on the position of 12 to 15 M\odot stars at the end of their nuclear life, in order to reconcile the theoretical tracks with the observed yellow supergiant progenitors. We perform calculations of 12 to 15 M\odot rotating stellar models using the Geneva stellar evolution code. To account for the uncertainties in the mass-loss rates during the RSG phase, we increase the mass-loss rate of the star (between 3 and 10 times the standard one) during that phase and compare the evolution of stars undergoing such high mass-loss rates with models computed with the standard mass-loss prescription. We show that the final position of the models in the Hertzsprung-Russell diagram depends on the mass loss they undergo during the red supergiant phase. With an increased mass-loss rate, we find that some models end their nuclear life at positions that are compatible with the observed position of several supernova progenitors. We conclude that an increased mass-loss rate (whom physical mechanism still need to be clarified) allows single star models to reproduce simultaneously the estimated position in the HRD of the YSG SN progenitors, as well as the SN type.