We report the discovery of the progenitor of the recent type IIn SN 2008S in the nearby galaxy NGC 6946. Surprisingly, it was not found in deep, pre-explosion optical images of its host galaxy taken with the Large Binocular Telescope, but only throug
h examination of archival Spitzer mid-IR data. A source coincident with the SN 2008S position is clearly detected in the 4.5, 5.8, and 8.0 micron IRAC bands, showing no evident variability in the three years prior to the explosion, yet is undetected at 3.6 and 24 micron. The distinct presence of ~440 K dust, along with stringent LBT limits on the optical fluxes, suggests that the progenitor of SN 2008S was engulfed in a shroud of its own dust. The inferred luminosity of 3.5x10^4 Lsun implies a modest mass of ~10 Msun. We conclude that objects like SN 2008S are not exclusively associated with the deaths or outbursts of very massive eta Carinae-like objects. This conclusion holds based solely on the optical flux limits even if our identification of the progenitor with the mid-IR source is incorrect.
In a variability survey of M81 using the Large Binocular Telescope we have discovered a peculiar eclipsing binary (MV ~ -7.1) in the field of the dwarf galaxy Holmberg IX. It has a period of 272 days and the light curve is well-fit by an overcontact
model in which both stars are overflowing their Roche lobes. It is composed by two yellow supergiants (V-I ~ 1 mag, T_eff = 4800 K), rather than the far more common red or blue supergiants. Such systems must be rare. While we failed to find any similar systems in the literature, we did, however note a second example. The SMC F0 supergiant R47 is a bright (MV ~ -7.5) periodic variable whose All Sky Automated Survey (ASAS) light curve is well-fit as a contact binary with a 181 day period. We propose that these systems are the progenitors of supernovae like SN 2004et and SN 2006ov, which appeared to have yellow progenitors. The binary interactions (mass transfer, mass loss) limit the size of the supergiant to give it a higher surface temperature than an isolated star at the same core evolutionary stage. We also discuss the possibility of this variable being a long-period Cepheid.
We investigate how the different types of supernovae are relatively affected by the metallicity of their host galaxy. We match the SAI Supernova Catalog to the SDSS-DR4 catalog of star-forming galaxies with measured metallicities. These supernova hos
t galaxies span a range of oxygen abundance from 12 + log(O/H) = 7.9 to 9.3 (~ 0.1 to 2.7 solar) and a range in absolute magnitude from MB = -15.2 to -22.2. To reduce the various observational biases, we select a subsample of well-characterized supernovae in the redshift range from 0.01 to 0.04, which leaves us with 58 SN II, 19 Ib/c, and 38 Ia. We find strong evidence that SN Ib/c are occurring in higher-metallicity host galaxies than SN II, while we see no effect for SN Ia relative to SN II. We note some extreme and interesting supernova-host pairs, including the metal-poor (~ 1/4 solar) host of the recent SN Ia 2007bk, where the supernova was found well outside of this dwarf galaxy. To extend the luminosity range of supernova hosts to even fainter galaxies, we also match all the historical supernovae with z < 0.3 to the SDSS-DR6 sky images, resulting in 1225 matches. This allows us to identify some even more extreme cases, such as the recent SN Ic 2007bg, where the likely host of this hypernova-like event has an absolute magnitude MB ~ -12, making it one of the least-luminous supernova hosts ever observed. This low-luminosity host is certain to be very metal poor (~ 1/20 solar), and therefore this supernova is an excellent candidate for association with an off-axis GRB. The two catalogs that we have constructed are available online and will be updated regularly. Finally, we discuss various implications of our findings for understanding supernova progenitors and their host galaxies.
We present extensive ugrizYHJK photometry and optical spectroscopy of SN 2005gj obtained by the SDSS-II and CSP Supernova Projects, which give excellent coverage during the first 150 days after the time of explosion. These data show that SN 2005gj is
the second clear case, after SN 2002ic, of a thermonuclear explosion in a dense circumstellar environment. Both the presence of singly and doubly ionized iron-peak elements (FeIII and weak SII, SiII) near maximum light as well as the spectral evolution show that SN 2002ic-like events are Type Ia explosions. Independent evidence comes from the exponential decay in luminosity of SN 2005gj, pointing to an exponential density distribution of the ejecta. The interaction of the supernova ejecta with the dense circumstellar medium is stronger than in SN 2002ic: (1) the supernova lines are weaker; (2) the Balmer emission lines are more luminous; and (3) the bolometric luminosity is higher close to maximum light. The velocity evolution of the Halpha components suggest that the CSM around SN 2005gj is clumpy and it has a flatter density distribution compared with the steady wind solution, in agreement with SN 2002ic. An early X-ray observation with Chandra gives an upper-limit on the mass loss rate from the companion of < 2x10^{-4} Msun/yr.
