No Arabic abstract
GCIRS 7, the brightest star in the Galactic central parsec, formed $6pm2$ Myr ago together with dozens of massive stars in a disk orbiting the central black-hole. It has been argued that GCIRS 7 is a pulsating body, on the basis of photometric variability. We present the first medium-resolution ($R=500$), K-band spectro-interferometric observations of GCIRS 7, using the GRAVITY instrument with the four auxiliary telescopes of the ESO VLTI. We looked for variations using two epochs, namely 2017 and 2019. We find GCIRS 7 to be moderately resolved with a uniform-disk photospheric diameter of $theta^*_text{UD}=1.55 pm 0.03$ mas ($R^*_text{UD}=1368 pm 26$ $R_odot$) in the K-band continuum. The narrow-band uniform-disk diameter increases above 2.3 $mu$m, with a clear correlation with the CO band heads in the spectrum. This correlation is aptly modeled by a hot ($T_text{L}=2368pm37$ K), geometrically thin molecular shell with a diameter of $theta_text{L}=1.74pm0.03$ mas, as measured in 2017. The shell diameter increased ($theta_text{L}=1.89pm0.03$ mas), while its temperature decreased ($T_text{L}=2140pm42$ K) in 2019. In contrast, the photospheric diameter $theta^*_text{UD}$ and the extinction up to the photosphere of GCIRS 7 ($A_{mathrm{K}_mathrm{S}}=3.18 pm 0.16$) have the same value within uncertainties at the two epochs. In the context of previous interferometric and photo-spectrometric measurements, the GRAVITY data allow for an interpretation in terms of photospheric pulsations. The photospheric diameter measured in 2017 and 2019 is significantly larger than previously reported using the PIONIER instrument ($theta_*=1.076 pm 0.093$ mas in 2013 in the H band). The parameters of the photosphere and molecular shell of GCIRS 7 are comparable to those of other red supergiants that have previously been studied using interferometry.
The stellar population in the central parsec of the Galaxy is dominated by an old (several Gyr) population, but young, massive stars dominate the luminosity function. We have studied the most luminous of these stars, GCIRS 7, in order to constrain the age of the recent star formation event in the Galactic Centre and to characterise it as an interferometric reference for observations of the Galactic Centre with the instrument GRAVITY, which will equip the Very Large Telescope Interferometer in the near future. We present the first H-band interferometric observations of GCIRS 7, obtained using the PIONIER visitor instrument on the VLTI using the four 8.2-m unit telescopes. In addition, we present unpublished K-band VLTI/AMBER data, build JHKL light-curves based on data spanning 4 decades, and measured the stars effective temperature using SINFONI spectroscopy. GCIRS 7 is marginally resolved at H-band (in 2013: uniform-disk diameter=1.076+/-0.093mas, R=960+/-92Rsun at 8.33+/-0.35kpc). We detect a significant circumstellar contribution at K-band. The star and its environment are variable in brightness and in size. The photospheric H-band variations are well modelled with two periods: P0~470+/-10 days (amplitude ~0.64mag) and long secondary period LSP~2700-2850 days (~1.1mag). As measured from CO equivalent width, <Teff>=3600+/-195K. The size, periods, luminosity (<Mbol>=-8.44+/-0.22) and effective temperature are consistent with an M1 supergiant with an initial mass of 22.5+/-2.5Msun and an age of 6.5-10Myr (depending on rotation). This age is in remarkable agreement with most estimates for the recent star formation event in the central parsec. Caution should be taken when using this star as an interferometric reference as it is variable in size, is surrounded by a variable circumstellar environment and large convection cells may form on its photosphere.
Context. Red supergiant stars possess surface features and extended molecular atmospheres. Photospheric convection may be a crucial factor of the levitation of the outer atmospheric layers. However, the mechanism responsible is still poorly understood. Aims. We image the stellar surface of V602 Carinae (V602 Car) to constrain the morphology and contrast of the surface features and of the extended atmospheric layers. Methods. We observed V602 Car with the Very Large Telescope Interferometer (VLTI) PIONIER instrument (1.53-1.78 $mathrm{mu}$m) between May and July 2016, and April and July 2019 with different telescope configurations. We compared the image reconstructions with 81 temporal snapshots of 3D radiative-hydrodynamics (RHD) CO$^5$BOLD simulations in terms of contrast and morphology, using the Structural Similarity Index. Results. The interferometric data are compatible with an overall spherical disk of angular diameter 4.4$pm$0.2 mas, and an extended molecular layer. In 2016, the reconstructed image reveals a bright arc-like feature toward the northern rim of the photospheric surface. In 2019, an arc-like feature is seen at a different orientation and a new peak of emission is detected on the opposite side. The contrasts of the reconstructed surface images are 11%$pm$2% and 9%$pm$2% for 2016 and 2019, respectively. The morphology and contrast of the two images are consistent with 3D RHD simulations, within our achieved spatial resolution and dynamic range. The extended molecular layer contributes 10--13% of the total flux with an angular diameter of 6--8 mas. It is present but not clearly visible in the reconstructed images because it is close to the limits of the achieved dynamic range. The presence of the molecular layer is not reproduced by the 3D RHD simulations. Conclusions...
We investigate the red supergiant (RSG) population of M31, obtaining radial velocities of 255 stars. These data substantiate membership of our photometrically-selected sample, demonstrating that Galactic foreground stars and extragalactic RSGs can be distinguished on the basis of B-V, V-R two-color diagrams. In addition, we use these spectra to measure effective temperatures and assign spectral types, deriving physical properties for 192 RSGs. Comparison with the solar-metallicity Geneva evolutionary tracks indicates astonishingly good agreement. The most luminous RSGs in M31 are likely evolved from 25-30 Mo stars, while the vast majority evolved from stars with initial masses of 20 Mo or less. There is an interesting bifurcation in the distribution of RSGs with effective temperatures that increases with higher luminosities, with one sequence consisting of early K-type supergiants, and with the other consisting of M-type supergiants that become later (cooler) with increasing luminosities. This separation is only partially reflected in the evolutionary tracks, although that might be due to the mis-match in metallicities between the solar Geneva models and the higher-than-solar metallicity of M31. As the luminosities increase the median spectral type also increases; i.e., the higher mass RSGs spend more time at cooler temperatures than do those of lower luminosities, a result which is new to this study. Finally we discuss what would be needed observationally to successfully build a luminosity function that could be used to constrain the mass-loss rates of RSGs as our Geneva colleagues have suggested.
We examine the problem of estimating the mass range corresponding to the observed red supergiant (RSG) progenitors of Type IIP supernovae. Using Monte Carlo simulations designed to reproduce the properties of the observations, we find that the approach of Davies & Beasor (2018) significantly overestimates the maximum mass, yielding an upper limit of Mh/Msun=20.5+/-2.6 for an input population with Mh/Msun=18. Our preferred Bayesian approach does better, with Mh/Msun=18.6+/-2.1 for the same input populations, but also tends to overestimate Mh. For the actual progenitor sample and a Salpeter initial mass function we find Mh/Msun=19.01-2.04+4.04 for the Eldridge et al. (2004) mass-luminosity relation used by Smartt et al. (2009) and Davies & Beasor (2018), and Mh/Msun=21.28_-2.28+4.52 for the Sukhbold et al. (2018) mass-luminosity relation. Based on the Monte Carlo simulations, we estimate that these are overestimated by 3.3+/-0.8Mh. The red supergiant problem remains.
We identify red supergiants (RSGs) in our spiral neighbors M31 and M33 using near-IR (NIR) photometry complete to a luminosity limit of log L/Lo=4.0. Our archival survey data cover 5 deg^2 of M31, and 3 deg^2 for M33, and are likely spatially complete for these massive stars. Gaia is used to remove foreground stars, after which the RSGs can be separated from asymptotic giant branch (AGB) stars in the color-magnitude diagram. The photometry is used to derive effective temperatures and bolometric luminosities via MARCS stellar atmosphere models. The resulting H-R diagrams show superb agreement with the evolutionary tracks of the Geneva evolutionary group. Our census includes 6400 RSGs in M31 and 2850 RSGs in M33 within their Holmberg radii; by contrast, only a few hundred RSGs are known so far in the Milky Way. Our catalog serves as the basis for a study of the RSG binary frequency being published separately, as well as future studies relating to the evolution of massive stars. Here we use the matches between the NIR-selected RSGs and their optical counterparts to show that the apparent similarity in the reddening of OB stars in M31 and M33 is the result of Malmquist bias; the average extinction in M31 is likely higher than that of M33. As expected, the distribution of RSGs follows that of the spiral arms, while the much older AGB population is more uniformly spread across each galaxys disk.