Using the properties of star forming regions in the host galaxies of
core-collapse supernovae (CCSNe), I extrapolate the properties of the
progenitor stars of these SNe. The goal is to determine what role progenitor
metallicity plays in the luminosity of a supernova by comparing the
environments of normal (average absolute magnitude of -17.5) and
super-luminous (absolute magnitude brighter than -20) supernovae.
First, I present spectroscopic oxygen abundance measurements of the hosts of
two super-luminous CCSNe, and show that, together with three other
measurements from the literature, a strong pattern emerges of super-luminous
SNe occurring in host galaxies that are metal-poor compared to the overall
galaxy population.
Next, I present metallicity measurements, host photometry, and best-fit galaxy
properties for a representative subsample of the Palomar Transient Factory's
first-year type II SN sample. The purpose of the in-depth study of this
sample is to serve as a reasonable standard of comparison for rare SN events.
The secondary aim is to continue to press the discussion in the community
about both the usefulness and the limitations of comparing SN hosts to samples
of galaxies. Because iron is a key source of opacity for massive star winds,
and at lower metallicity, $\alpha$-elements like oxygen are enhanced relative
to iron compared to the solar mixture, I present a conversion from gas-phase
oxygen abundance measurements to implied iron abundances, and discuss the
logic, uses, and limitations of this conversion.
Finally, I present metallicity measurements for a sample of eight hosts of
super-luminous CCSNe. Comparing them and four measurements from the
literature, I find that super-luminous CCSNe come from much more metal-poor
environments than normal type II SNe. This implies that the progenitors of
super-luminous CCSNe are much poorer in iron than the progenitors of
normal type II SNe, suggesting mass loss is a key factor for abnormally high
luminosity.