Context. Observationally, supernovae (SNe) are divided into subclasses according to their distinct characteristics. This diversity naturally reflects the diversity in the progenitor stars. It is not entirely clear, however, how different evolutionary paths leading massive stars to become an SN are governed by fundamental parameters such as progenitor initial mass and metallicity.
Aims. This paper places constraints on progenitor initial mass and metallicity in distinct core-collapse SN subclasses through a study of the parent stellar populations at the explosion sites.
Methods. Integral field spectroscopy (IFS) of 83 nearby SN explosion sites with a median distance of 18 Mpc has been collected and analysed, enabling detection and spectral extraction of the parent stellar population of SN progenitors. From the parent stellar population spectrum, the initial mass and metallicity of the coeval progenitor are derived by means of comparison to simple stellar population models and strong-line methods. Additionally, near-infrared IFS was employed to characterise the star formation history at the explosion sites.
Results. No significant metallicity differences are observed among distinct SN types. The typical progenitor mass is found to be highest for SN type Ic, followed by type Ib, then types IIb and II. Type IIn is the least associated with young stellar populations and thus massive progenitors. However, statistically significant differences in progenitor initial mass are observed only when comparing SNe IIn with other subclasses. Stripped-envelope SN progenitors with initial mass estimates lower than 25 M⊙ are found; they are thought to be the result of binary progenitors. Confirming previous studies, these results support the notion that core-collapse SN progenitors cannot arise from single-star channels only, and both single and binary channels are at play in the production of core-collapse SNe. Near-infrared IFS suggests that multiple stellar populations with different ages may be present in some of the SN sites. As a consequence, there could be a non-negligible amount of contamination from old populations, and therefore the individual age estimates are effectively lower limits.