The Wolf-Rayet Content of M33

TL;DR

This comprehensive survey of Wolf-Rayet stars in M33 reveals a higher WC/WN ratio in the galaxy's center than models predict, providing insights into metallicity effects on stellar evolution.

Contribution

The study presents a complete, sensitive survey of M33's Wolf-Rayet stars, improving spectral classifications and revealing a galactic metallicity gradient's impact on WR populations.

Findings

206 confirmed WRs in M33, complete to ~5%

Higher WC/WN ratio in central regions than models predict

Supports a larger oxygen abundance gradient in M33

Abstract

Wolf-Rayet stars (WRs) are evolved massive stars, and the relative number of WC-type and WN-type WRs should vary with metallicity, providing a sensitive test of stellar evolutionary theory. The observed WC/WN ratio is much higher than that predicted by theory in some galaxies but this could be due to observational incompleteness for WN-types, which have weaker lines. Previous studies of M33's WR content show a galactocentric gradient in the relative numbers of WCs and WNs, but only small regions have been surveyed with sufficient sensitivity to detect all of the WNs. Here we present a sensitive survey for WRs covering all of M33, finding 55 new WRs, mostly of WN type. Our spectroscopy also improves the spectral types of many previously known WRs, establishing in one case that the star is actually a background quasar. The total number of spectroscopically confirmed WRs in M33 is 206, a number we argue is complete to approximately 5%, with most WRs residing in OB associations, although approximately 2% are truly isolated. The WC/WN ratio in the central regions (<2 kpc) of M33 is much higher than that predicted by the current Geneva evolutionary models, while the WC/WN ratios in the outer regions are in good accord, as are the values in the SMC and LMC. The WC/WN ratio and the WC subtype distribution both argue that the oxygen abundance gradient in M33 is significantly larger than found by some recent studies, but are consistent with the two-component model proposed by Magrini et al.