Context. Massive star evolution at low metallicity is closely connected to many fields in high-redshift astrophysics, but is poorly understood so far. Because of its metallicity of ∼0.2 Z⊙, its proximity, and because it is currently forming stars, the Small Magellanic Cloud (SMC) is a unique laboratory in which to study metal-poor massive stars.
Aims. We seek to improve the understanding of this topic using available SMC data and a comparison to stellar evolution predictions.
Methods. We used a recent catalog of spectral types in combination with Gaia magnitudes to calculate temperatures and luminosities of bright SMC stars. By comparing these with literature studies, we tested the validity of our method, and using Gaia data, we estimated the completeness of stars in the catalog as a function of luminosity. This allowed us to obtain a nearly complete view of the most luminous stars in the SMC. We also calculated the extinction distribution, the ionizing photon production rate, and the star formation rate.
Results. Our results imply that the SMS hosts only ∼30 very luminous main-sequence stars (M ≥ 40 M⊙; L ≳ 3 ⋅ 105 L⊙), which are far fewer than expected from the number of stars in the luminosity range 3 ⋅ 104 < L/L⊙ < 3 ⋅ 105 and from the typically quoted star formation rate in the SMC. Even more striking, we find that for masses above M ≳ 20 M⊙, stars in the first half of their hydrogen-burning phase are almost absent. This mirrors a qualitatively similar peculiarity that is known for the Milky Way and Large Magellanic Cloud. This amounts to a lack of hydrogen-burning counterparts of helium-burning stars, which is more pronounced for higher luminosities. We derived the H I ionizing photon production rate of the current massive star population. It agrees with the H α luminosity of the SMC.
Conclusions. We argue that a declining star formation rate or a steep initial mass function are unlikely to be the sole explanations for the dearth of young bright stars. Instead, many of these stars might be embedded in their birth clouds, although observational evidence for this is weak. We discuss implications for the role that massive stars played in cosmic reionization, and for the top end of the initial mass function.