There are currently two optical interferometry recombiners that can provide spectral resolutions better than 10000, AMBER/VLTI operating in the H-K bands, and VEGA/CHARA, recently commissioned, operating in the visible. These instruments are well sui
ted to study the wind activity of the brightest AB supergiants in our vicinity, in lines such as H$alpha$ or BrGamma. We present here the first observations of this kind, performed on Rigel (B8Ia) and Deneb (A2Ia). Rigel was monitored by AMBER in two campaigns, in 2006-2007 and 2009-2010, and observed in 2009 by VEGA; whereas Deneb was monitored in 2008-2009 by VEGA. The extension of the Halpha and BrGamma line forming regions were accurately measured and compared with CMFGEN models of both stars. Moreover, clear signs of activity were observed in the differential visibility and phases. These pioneer observations are still limited, but show the path for a better understanding of the spatial structure and temporal evolution of localized ejections using optical interferometry.
BA-type supergiants are amongst the most optically-bright stars. They are observable in extragalactic environments, hence potential accurate distance indicators. Emission activity in the Halpha line of the BA supergiants Rigel (B8Ia) and Deneb (A2Ia)
is indicative of presence of localized time-dependent mass ejections. Here, we employ optical interferometry to study the Halpha line-formation region in these stellar environments. High spatial- (0.001 arcsec) and spectral- (R=30 000) resolution observations of Halpha were obtained with the visible recombiner VEGA installed on the CHARA interferometer, using the S1S2 array-baseline (34m). Six independent observations were done on Deneb over the years 2008 and 2009, and two on Rigel in 2009. We analyze this dataset with the 1D non-LTE radiative-transfer code CMFGEN, and assess the impact of the wind on the visible and near-IR interferometric signatures, using both Balmer-line and continuum photons. We observe a visibility decrease in Halpha for both Rigel and Deneb, suggesting that the line-formation region is extended (1.5-1.75 R*). We observe a significant visibility decrease for Deneb in the SiII6371 line. We witness time variations in the differential phase for Deneb, implying an inhomogeneous and unsteady circumstellar environment, while no such variability is seen in differential visibilities. Radiative-transfer modeling of Deneb, with allowance for stellar-wind mass loss, accounts fairly well for the observed decrease in the Halpha visibility. Based on the observed differential visibilities, we estimate that the mass-loss rate of Deneb has changed by less than 5%.
We present a multi-epoch quantitative spectroscopic analysis of the Type IIn SN 1994W, an event interpreted by Chugai et al. as stemming from the interaction between the ejecta of a SN and a 0.4Msun circumstellar shell ejected 1.5yr before core colla
pse. During the brightening phase, our models suggest that the source of optical radiation is not unique, perhaps associated with an inner optically-thick Cold Dense Shell (CDS) and outer optically-thin shocked material. During the fading phase, our models support a single source of radiation, an hydrogen-rich optically-thick layer with a near-constant temperature of ~7000K that recedes from a radius of 4.3x10^15 at peak to 2.3x10^15cm 40 days later. We reproduce the hybrid narrow-core broad-wing line profile shapes of SN 1994W at all times, invoking an optically-thick photosphere exclusively (i.e., without any external optically-thick shell). In SN 1994W, slow expansion makes scattering with thermal electrons a key escape mechanism for photons trapped in optically-thick line cores, and allows the resulting broad incoherent electron-scattering wings to be seen around narrow line cores. In SNe with larger expansion velocities, the thermal broadening due to incoherent scattering is masked by the broad profile and the dominant frequency redshift occasioned by bulk motions. Given the absence of broad lines at all times and the very low 56Ni yields, we speculate whether SN 1994W could have resulted from an interaction between two ejected shells without core collapse. The high conversion efficiency of kinetic to thermal energy may not require a SN-like energy budget for SN1994W.
We report the serendipitous detection by GALEX of fast (<1 day) rising (>1 mag) UV emission from two Type II plateau (II-P) supernovae (SNe) at z=0.185 and 0.324 discovered by the Supernova Legacy Survey. Optical photometry and VLT spectroscopy 2 wee
ks after the GALEX detections link the onset of UV emission to the time of shock breakout. Using radiation hydrodynamics and non-LTE radiative transfer simulations, and starting from a standard red supergiant (RSG; Type II-P SN progenitor) star evolved self-consistently from the main sequence to iron core collapse, we model the shock breakout phase and the 55 hr that follow. The small scale height of our RSG atmosphere model suggests that the breakout signature is a thermal soft X-ray burst (lambda_peak ~ 90AA) with a duration of <~ 2000 s. Longer durations are possible but require either an extended and tenuous non-standard envelope, or an unusually dense RSG wind with dot{M} ~ 10^(-3) Msun yr^(-1). The GALEX observations miss the peak of the luminous (M_FUV ~ -20) UV burst but unambiguously capture the rise of the emission and a subsequent 2 day long plateau. The postbreakout, UV-bright plateau is a prediction of our model in which the shift of the peak of the spectral energy distribution (SED) from ~100 to ~1000AA and the ejecta expansion both counteract the decrease in bolometric luminosity from ~10^11 to ~10^9 L_sun over that period. Based on the observed detection efficiency of our study we make predictions for the breakout detection rate of the GALEX Time Domain Survey.
Spectroscopic modeling of Type II supernovae (SNe) generally assumes steady-state. Following the recent suggestion of Utrobin & Chugai, but using the 1D non-LTE line-blanketed model atmosphere code CMFGEN, we investigate the effects of including time
-dependent terms that appear in the statistical and radiative equilibrium equations. We base our discussion on the ejecta properties and the spectroscopic signatures obtained from time-dependent simulations, investigating different ejecta configurations, and covering their evolution from one day to six weeks after shock breakout. Compared to equivalent steady-state models, our time-dependent models produce SN ejecta that are systematically over-ionized, affecting helium at one week after explosion, but ultimately affecting all ions after a few weeks. While the continuum remains essentially unchanged, time-dependence effects on observed spectral lines are large. At the recombination epoch, HI lines and NaID are considerably stronger and broader than in equivalent steady-state models, while CaII8500A is weakened. If time dependence is allowed for, the HeI lines at 5875A and 10830A appear about 3 times stronger at one week, and HeI10830A persists as a blue-shifted absorption feature even at 6 weeks after explosion. Time dependence operates through the energy gain from changes in ionization and excitation, and, perhaps more universally across SN types, from the competition between recombination and expansion, which in-turn, can be affected by optical-depth effects. Our time-dependent models compare well with observations of the low-luminosity low-velocity SN 1999br and the more standard SN 1999em, reproducing the Halpha line strength at the recombination epoch, and without the need for setting unphysical requirements on the magnitude of nickel mixing.
