Supernovae arise from progenitor stars occupying the upper end of the initial mass function. Their extreme brightness allows individual massive stars to be detected at cosmic distances, lending supernovae great potential as tracers of the upper end o
f the IMF and its evolution. Exploiting this potential requires progress in many areas of supernova science. These include understanding the progenitor masses that produce various types of supernovae and accurately characterizing the supernova outburst and the environment in which it was produced. I present some preliminary work identifying the environmental conditions that produce the most luminous supernovae, believed to arise from stars with masses greater than 100 M_sun. I illustrate that the presence of these extreme supernovae in small star-forming dwarfs can be used to test our understanding of the upper end of the IMF.
We use GALEX ultraviolet (UV) and optical integrated photometry of the hosts of seventeen luminous supernovae (LSNe, having peak M_V < -21) and compare them to a sample of 26,000 galaxies from a cross-match between the SDSS DR4 spectral catalog and G
ALEX interim release 1.1. We place the LSNe hosts on the galaxy NUV-r versus M_r color magnitude diagram (CMD) with the larger sample to illustrate how extreme they are. The LSN hosts appear to favor low-density regions of the galaxy CMD falling on the blue edge of the blue cloud toward the low luminosity end. From the UV-optical photometry, we estimate the star formation history of the LSN hosts. The hosts have moderately low star formation rates (SFRs) and low stellar masses (M_*) resulting in high specific star formation rates (sSFR). Compared with the larger sample, the LSN hosts occupy low-density regions of a diagram plotting sSFR versus M_* in the area having higher sSFR and lower M_*. This preference for low M_*, high sSFR hosts implies the LSNe are produced by an effect having to do with their local environment. The correlation of mass with metallicity suggests that perhaps wind-driven mass loss is the factor that prevents LSNe from arising in higher-mass, higher-metallicity hosts. The massive progenitors of the LSNe (>100 M_sun), by appearing in low-SFR hosts, are potential tests for theories of the initial mass function that limit the maximum mass of a star based on the SFR.
We use multi-wavelength, matched aperture, integrated photometry from GALEX, SDSS and the RC3 to estimate the physical properties of 166 nearby galaxies hosting 168 well-observed Type Ia supernovae (SNe Ia). Our data corroborate well-known features t
hat have been seen in other SN Ia samples. Specifically, hosts with active star formation produce brighter and slower SNe Ia on average, and hosts with luminosity-weighted ages older than 1 Gyr produce on average more faint, fast and fewer bright, slow SNe Ia than younger hosts. New results include that in our sample, the faintest and fastest SNe Ia occur only in galaxies exceeding a stellar mass threshhold of ~10^10 M_sun, indicating that their progenitors must arise in populations that are older and/or more metal rich than the general SN Ia population. A low host extinction sub-sample hints at a residual trend in peak luminosity with host age, after correcting for light-curve shape, giving the appearance that older hosts produce less-extincted SNe Ia on average. This has implications for cosmological fitting of SNe Ia and suggests that host age could be useful as a parameter in the fitting. Converting host mass to metallicity and computing 56Ni mass from the supernova light curves, we find that our local sample is consistent with a model that predicts a shallow trend between stellar metallicity and the 56Ni mass that powers the explosion, but we cannot rule out the absence of a trend. We measure a correlation between 56Ni mass and host age in the local universe that is shallower and not as significant as that seen at higher redshifts. The details of the age -- 56Ni mass correlations at low and higher redshift imply a luminosity-weighted age threshhold of ~3 Gyr for SN Ia hosts, above which they are less likely to produce SNe Ia with 56Ni masses above ~0.5 M_sun. (Abridged)
We exploit the accumulating, high-quality, multi-wavelength imaging data of nearby supernova (SN) hosts to explore the relationship between SN production and host galaxy evolution. The Galaxy Evolution Explorer (GALEX, Martin et al., 2005) provides u
ltraviolet (UV) imaging in two bands, complementing data in the optical and infra-red (IR). We compare host properties, derived from spectral energy distribution (SED) fitting, with nearby, well-observed SN Ia light curve properties. We also explore where the hosts of different types of SNe fall relative to the red and blue sequences on the galaxy UV-optical color-magnitude diagram (CMD, Wyder et al., 2007). We conclude that further exploration and larger samples will provide useful results for constraining the progenitors of SNe.
We quantify the effect of supernova Type Ia peculiar velocities on the derivation of cosmological parameters. The published distant and local Ia SNe used for the Supernova Legacy Survey first-year cosmology report form the sample for this study. Whil
e previous work has assumed that the local SNe are at rest in the CMB frame (the No Flow assumption), we test this assumption by applying peculiar velocity corrections to the local SNe using three different flow models. The models are based on the IRAS PSCz galaxy redshift survey, have varying beta = Omega_m^0.6/b, and reproduce the Local Group motion in the CMB frame. These datasets are then fit for w, Omega_m, and Omega_Lambda using flatness or LambdaCDM and a BAO prior. The chi^2 statistic is used to examine the effect of the velocity corrections on the quality of the fits. The most favored model is the beta=0.5 model, which produces a fit significantly better than the No Flow assumption, consistent with previous peculiar velocity studies. By comparing the No Flow assumption with the favored models we derive the largest potential systematic error in w caused by ignoring peculiar velocities to be Delta w = +0.04. For Omega_Lambda, the potential error is Delta Omega_Lambda = -0.04 and for Omega_m, the potential error is Delta Omega_m < +0.01. The favored flow model (beta=0.5) produces the following cosmological parameters: w = -1.08 (+0.09,-0.08), Omega_m = 0.27 (+0.02,-0.02) assuming a flat cosmology, and Omega_Lambda = 0.80 (+0.08,-0.07) and Omega_m = 0.27 (+0.02,-0.02) for a w = -1 (LambdaCDM) cosmology.
