Unveiling Massive Main-Sequence Stars in Sextans A through Panchromatic Photometry

Abstract

We present a study of the metal-poor (~6% Z_sun) massive (>8 M_sun) main-sequence star population in the star-forming dwarf galaxy Sextans A. By modeling near-UV to near-IR photometry of individual stars using the Bayesian Extinction and Stellar Tool (BEAST) we infer stellar parameters such as effective temperature, luminosity, and initial mass. We identify 867 massive main-sequence star candidates (present-day mass >8 M_sun and surface gravity >3.7 dex [cgs]) with a plausible spectral energy distribution (SED) fit, 500 of which show a probable SED fit. Comparisons to spectral types of existing observed spectra are consistent with the BEAST-derived stellar parameters, with most discrepancies explained. We identify 292 OBe star candidates through IR photometric signatures and find lower-limit OBe fractions of 15% for M > 8 M_sun, 23% for M > 15 M_sun, and 17% for M > 20 M_sun. We find 57 OB associations and that 24-28% of massive stars are isolated (distance to nearest massive star >28 pc). We discuss six likely runaway candidates (suggested velocities of ~ 50-340 km/s) not clearly associated with any major star-forming complexes. Lastly, we predict Lyman continuum (LyC) escape fractions of f_esc=0.27-0.76 across the star-forming regions and a global value of 0.35-0.71 by assuming low overall extinction and a range of porous geometries, indicating efficient leakage of ionizing photons. Future spectroscopic follow-up and resolved ISM studies will refine these constraints and solidify Sextans A as a benchmark for studying massive-star evolution and feedback at extremely low metallicity.

Figures (3)

• Figure 1: Optical image (left: color, right:grayscale) of Sextans A obtained with the Nicholas U. Mayall 4-m Telescope at Kitt Peak National Observatory. Left: We indicate the directions of North and East, along with a distance scale of 500 pc. Middle: Footprints of the HST imaging produced by the LUVIT Survey Gilbert25 are overlaid on the optical image. The turquoise footprint shows the HST UV observations, the green and black footprint shows the HST optical observations, and the red footprint denotes the HST IR footprint. (credit: KPNO/NOIRLab/NSF/AURA Data obtained and processed by: P. Massey (Lowell Obs.), G. Jacoby, K. Olsen, & C. Smith (AURA/NSF) Image processing: T.A. Rector (University of Alaska Anchorage/NSF NOIRLab), M. Zamani (NSF NOIRLab) & D. de Martin (NSF NOIRLab) Right: Coordinates of resolved sources within the footprint, color-coded by the number of observed sources.
• Figure 2: Select HST-based UV-optical (top) and optical-only (bottom) CMDs of Sextans A. We overplot select MIST evolutionary tracks for massive stars, assuming [Fe/H] =$-$1.22, v/vcrit=0.4, and A$_V =0.112$. We indicate the typical errors for each CMD in the left-hand of each subplot.
• Figure 3: Example SED fits in each category of $\chi^2$ regimes: probable (top left), plausible (top right), plausible marked as OBe candidate (bottom left), and bad (lower right). The data are shown as black points. The median fit stellar (pink), stellar$+$dust (purple), and stellar$+$dust$+$bias (cyan) are overplotted. The sub-panels show the residuals in units of $\sigma$. The top left panel shows all residuals of the best model (cyan) within 1.5$\sigma$. The top right panel shows residuals of the best model (cyan) for all UV+optical filters within 3 $\sigma$; only the IR filter HST/F160W shows a large discrepancy and lies within 9 $\sigma$. The bottom left panel shows residuals of the best model (cyan) within 3 $\sigma$, except for the HST/F814W filter, where the residual is within 6 $\sigma$. This star is identified as an OBe star candidate, which is likely why HST/F814W is not well-captured by the underlying model that does not include a disk component. The bottom right panel shows the largest deviations in the residual with the best model (cyan) being within 22 $\sigma$.