Search

We test the viability of training machine learning algorithms with synthetic H alpha line profiles to determine the inclination angles of Be stars (the angle between the central B star's rotation axis and the observer's line of sight) from a single observed medium-resolution, moderate S/N, spectrum. The performance of three different machine learning algorithms were compared: neural networks tasked with regression, neural networks tasked with classification, and support vector regression. Of these three algorithms, neural networks tasked with regression consistently outperformed the other methods with a RMSE error of 7.6 degrees on an observational sample of 92 galactic Be stars with inclination angles known from direct H alpha profile fitting, from the spectroscopic signature of gravitational darkening, and, in a few cases, from interferometric observations that resolved the disk. The trained neural networks enable a quick and useful determination of the inclination angles of observed Be stars which can be used to search for correlated spin axes in young open clusters or to extract an equatorial rotation velocity from a measurement of v sin(i)., Comment: 14 pages, 7 figures

Using a sample of 92 Galactic Be stars, we compare inclination angles (the angle between a star's rotation axis and the line-of-sight) determined from H$\alpha$ emission line profile fitting to those determined by the spectroscopic signature of gravitational darkening. We find good agreement: 70% of the sample (64 out of 92 stars) is consistent with zero difference between the two methods using $1\sigma$ errors, and there is a strong linear correlation coefficient between the two methods of $r=+0.63\pm0.05$. There is some evidence that the H$\alpha$ profile fitting method overestimates the inclination angle for $i\lesssim 25^\circ$, perhaps due to the neglect of incoherent electron scattering on the H$\alpha$ line widths, while the gravitational darkening method underestimates the inclination angle for $i\gtrsim70^\circ$, perhaps due to the neglect of disk radiative transfer effects on the optical spectrum. Overall, it is demonstrated that a single H$\alpha$ spectrum of modest resolution and SNR can be used to extract a useful estimate for the inclination angle of an individual Be star. This allows equatorial rotation velocities for individual Be stars to be derived from $v \sin i $ measurements and will allow Be stars to be used to search for correlated spin axes in young, open clusters if unbiased (with respect to inclination) samples of Be stars are used., Comment: 21 pages, 18 figures

We demonstrate that the angle between star's rotation axis and the observer's line-of-sight, usually called the inclination angle, can be reliably determined for Be stars via H$\alpha$ emission-line profile fitting. We test our method on a sample of 11 Be stars with available long-baseline interferometric data from the Navy Precision Optical Interferometer~(NPOI). We fit the H$\alpha$ emission line profile of each star to obtain a spectroscopic inclination angle $i_{\rm H\alpha}$. We then obtain an independent inclination angle estimate, $i_{\rm V^2}$, by fitting the observed interferometric visibilities with model visibilities based on a purely geometric representation of the light distribution on the sky. The sample differences, $\Delta i \equiv i_{\rm H\alpha} - i_{\rm V^2}$, are normally distributed with a mean of zero and a standard deviation of $6.7$ degrees, and the linear correlation coefficient between $i_{\rm H\alpha}$ and $i_{\rm V^2}$ is $r=0.93$. As Be stars comprise upwards of one fifth of all main-sequence B-type stars, this H$\alpha$ line profile fitting technique has the potential to provide an efficient method for detecting correlated stellar spin axes in young open clusters. Furthermore, if the orientation of the Be star circumstellar disk on the plane of the sky can be constrained by polarization measurements, it is possible to determine the full 3D stellar rotation vector of each Be star., Comment: 19 pages, 23 figures, accepted for publication in The Astrophysical Journal

We present results from a 2-epoch HST H$\alpha$ emission line survey of the Andromeda Galaxy that overlaps the footprint of the Panchromatic Hubble Andromeda Treasury (PHAT) survey. We find 552 (542) classical Be stars and 8429 (8556) normal B-type stars in epoch # 1 (epoch # 2), yielding an overall fractional Be content of 6.15% $\pm$0.26% (5.96% $\pm$0.25%). The fractional Be content decreased with spectral sub-type from $\sim$23.6% $\pm$2.0% ($\sim$23.9% $\pm$2.0%) for B0-type stars to $\sim$3.1% $\pm$0.34% ($\sim$3.4% $\pm$0.35%) for B8-type stars in epoch # 1 (epoch # 2). We observe a clear population of cluster Be stars at early fractional main sequence lifetimes, indicating that a subset of Be stars emerge onto the ZAMS as rapid rotators. Be stars are 2.8x rarer in M31 for the earliest sub-types compared to the SMC, confirming that the fractional Be content decreases in significantly more metal rich environments (like the Milky Way and M31). However, M31 does not follow a clear trend of Be fraction decreasing with metallicity compared to the Milky Way, which may reflect that the Be phenomenon is enhanced with evolutionary age. The rate of disk-loss or disk-regeneration episodes we observe, 22% $\pm$ 2% yr$^{-1}$, is similar to that observed for seven other Galactic clusters reported in the literature, assuming these latter transient fractions scale by a linear rate. The similar number of disk-loss events (57) as disk-renewal events (43) was unexpected since disk dissipation time-scales can be $\sim$2x the typical time-scales for disk build-up phases., Comment: Accepted to AJ

Through the modelling of near-infrared photometry of star plus disk systems with the codes BEDISK/BERAY, we successfully describe the WISE photometric characteristics of Be stars in five young open clusters, NGC 663, NGC 869, NGC 884, NGC 3766 and NGC 4755, broadly studied in the literature. WISE photometry allows previously known Be stars to be detected and to find new Be candidates which could be confirmed spectroscopically. The location of Be stars in the WISE colour-magnitude diagram, separates them in two groups; active (Be stars hosting a developed circumstellar disk) and quiescent objects (Be stars in a diskless phase), and this way, we can explore how often stars are observed in these different stages. The variability observed in most active variable Be stars is compatible with a disk dissipation phase. We find that 50 percent of Be stars in the studied open clusters are in an active phase. We can interpret this as Be stars having a developed circumstellar disk half of the time. The location of Be stars with a developed disk in the CMD require mass loss rates in agreement with values recently reported in the literature. For these objects, we expect to have a tight relation between the equivalent width of the H{\alpha} line and the mass of the disk, if the inclination is known. Also, near-infrared photometry of Be stars in stellar clusters has the potential of being useful to test whether there is a preferential viewing angle., Comment: 25 pages, 14 figures, Accepted for publication in AJ

Photospheric nitrogen abundances for a sample of 26 Be stars and 16 normal B stars were found using high-resolution spectra from the Magnetism in Massive Stars (MiMeS) spectroscopic survey. Nitrogen abundances were obtained using non-LTE equivalent widths and line profiles, and Monte Carlo simulation was used to determine the error bounds of the measured nitrogen abundances due to uncertainties in the adopted stellar parameters, continuum normalization, and atomic data. In addition, the effects of the gravitational darkening and disk contamination on the measured Be star nitrogen abundances were investigated. About one third of the sample Be stars and half of the normal B-type stars have nitrogen enrichments that may be due to rotational mixing in these rapidly rotating objects. Corrections for gravitational darkening and disk contamination do not significantly change the overall results for the Be star sample. The average nitrogen abundance of the B and Be star samples coincide with the solar abundance, although the dispersion in the nitrogen abundances is much larger in the Be star sample. The Be star sample also has a significant fraction of sub-solar nitrogen abundance objects which are not present in significant numbers in the B star sample. This may point to yet unresolved systematic errors in the analysis of the Be stars, Comment: 24 pages; 27 figures, Accepted for publication in the MNRAS

We utilize a multi-step modelling process to produce synthetic interferometric and spectroscopic observables, which are then compared to their observed counterparts. Our extensive set of interferometric observations of the Be star 48 Per, totaling 291 data points, were obtained at the Navy Precision Optical Interferometer from 2006 Nov 07 to 2006 Nov 23. Our models were further constrained by comparison with contemporaneous H$\alpha$ line spectroscopy obtained at the John S. Hall Telescope at Lowell Observatory on 2006 Nov 1. Theoretical spectral energy distributions, SEDs, for 48~Per were confirmed by comparison with observations over a wavelength regime of 0.4 to 60 microns from Touhami et al. (2010) and Vieira et al. (2017). Our best-fitting combined model from H$\alpha$ spectroscopy, H$\alpha$ interferometry and SED fitting has a power law density fall off, $n$, of 2.3 and an initial density at the stellar surface of $\rho_0 = 1.0 \times 10^{-11} \rm{g\,cm}^{-3}$ with a inclination constrained by H$\alpha$ spectroscopy and interferometry of 45$^{\circ} \pm 5^{\circ}$. The position angle for the system, measured east from north, is 121$^{\circ}$ $\pm$ 1$^{\circ}$. Our best-fit model shows that the disk emission originates in a moderately large disk with a radius of 25 R$_{*}$ which is consistent with a disk mass of approximately 5 $\times$ 10$^{24}$ g or 3 $\times$ 10$^{-10}$ M$_{*}$. Finally, we compare our results with previous studies of 48~Per by Quirrenbach et al. (1997) with agreement but find, with our much larger data set, that our disk size contradicts the findings of Delaa et al. (2011)., Comment: 26 pages, 9 figures

The circumstellar disk density distributions for a sample of 63 Be southern stars from the BeSOS survey were found by modelling their H$\alpha$ emission line profiles. These disk densities were used to compute disk masses and disk angular momenta for the sample. Average values for the disk mass are 3.4$\times$10$^{-9}$ and 9.5$\times$10$^{-10}$ $M_{\star}$ for early (B0-B3) and late (B4-B9) spectral types, respectively. We also find that the range of disk angular momentum relative to the star are between 150-200 and 100-150 $J_{\star}/M_{\star}$, again for early and late-type Be stars respectively. The distributions of the disk mass and disk angular momentum are different between early and late-type Be stars at a 1\% level of significance. Finally, we construct the disk mass distribution for the BeSOS sample as a function of spectral type and compare it to the predictions of stellar evolutionary models with rapid rotation. The observed disk masses are typically larger than the theoretical predictions, although the observed spread in disk masses is typically large.

We investigate the physical properties of the inner gaseous disks of the three, hot, Herbig B2e stars, HD 76534, HD 114981 and HD 216629, by modelling CFHT-ESPaDOns spectra using non-LTE radiative transfer codes. We assume that the emission lines are produced in a circumstellar disk heated solely by the photospheric radiation from the central star in order to test if the optical and near-IR emission lines can be reproduced without invoking magnetospheric accretion. The inner gaseous disk density was assumed to follow a simple power-law in the equatorial plane, and we searched for models that could reproduce observed lines of HI (H$\alpha$ and H$\beta$), HeI, CaII and FeII. For the three stars, good matches were found for all emission line profiles individually; however, no density model based on a single power-law was able to reproduce all of the observed emission lines. Among the single power-law models, the one with the gas density varying as $\sim 10^{-10}\,(R_{*}/R)^3\, \rm g\,cm^{-3}$ in the equatorial plane of a 25 R$_{*}$ ($0.78$ AU) disk did the best overall job of representing the optical emission lines of the three stars. This model implies a mass for the H$\alpha$-emitting portion of the inner gaseous disk of $\sim 10^{-9} M_*$. We conclude that the optical emission line spectra of these HBe stars can be qualitatively reproduced by a $\approx\,1$ AU, geometrically thin, circumstellar disk of negligible mass compared to the central star in Keplerian rotation and radiative equilibrium., Comment: Accepted to be published in AAS Journals. 30 pages, 24 figures and 6 tables

We present eleven years of high-resolution, spectroscopic observations for the star EM~Cep. EM~Cep switches between B and Be star states, as revealed by the level of H$\alpha$ emission, but spends most of its time in the B~star state. EM~Cep has been considered to be an eclipsing, near contact binary of nearly equal-massed B stars in order to reproduce regular photometric variations; however, this model is problematic due to the lack of any observed Doppler shift in the spectrum. Our observations confirm that there are no apparent Doppler shifts in the wide spectral lines H$\alpha$ and HeI~6678 in either the B or Be star states.The profiles of HeI~6678 typically exhibited a filled-in absorption core, but we detected weak emission in this line during the highest Be state. Given the lack of observed Doppler shifts, we model EM~Cep as an isolated Be star with a variable circumstellar disk. We can reproduce the observed H$\alpha$ emission profiles over the eleven year period reasonably well with disk masses on the order of $3$ to $10\times\,10^{-11}\;M_*$ in the Be state with the circumstellar disk seen at an inclination of $78^{\circ}$ to the line of sight. From a disk ejection episode in 2014, we estimate a mass loss rate of $\approx\,3\times 10^{-9}\;\rm M_{\odot}\,yr^{-1}$. The derived disk density parameters are typical of those found for the classical Be stars. We therefore suggest that the EM~Cep is a classical Be star and that its photometric variations are the result of $\beta\;$Cep or non-radial pulsations., Comment: 15 pages, 14 figures

We present a study of FeII emission in the near-infrared region (NIR) for 25 active galactic nuclei (AGNs) to obtain information about the excitation mechanisms that power it and the location where it is formed. We employ a NIR FeII template derived in the literature and found that it successfully reproduces the observed FeII spectrum. The FeII bump at 9200 Angstroms detected in all objects studied confirms that Lyman-alpha fluorescence is always present in AGNs. The The correlation found between the flux of the 9200 Angstroms bump, the 1 micron lines and the optical FeII imply that Lyman-alpha fluorescence plays an important role in the FeII production. We determined that at least 18% of the optical FeII is due to this process while collisional excitation dominates the production of the observed FeII. The line profiles of FeII 10502, OI 11287, CaII 8664 and Paschen-beta were compared to gather information about the most likely location where they are emitted. We found that FeII, OI and CaII have similar widths and are, on average, 30% narrower than Paschen-beta. Assuming that the clouds emitting the lines are virialized, we show that the FeII is emitted in a region twice as far from the central source than Paschen-beta. The distance though strongly varies: from 8.5 light-days for NGC4051 to 198.2 light-days for Mrk509. Our results reinforce the importance of the FeII in the NIR to constrain critical parameters that drive its physics and the underlying AGN kinematics as well as more accurate models aimed at reproducing this complex emission., Comment: Accepted for publication in The Astrophysical Journal

We attempt to constrain the physical properties of the inner, gaseous disk of the Herbig Be star BD+65 1637 using non-LTE, circumstellar disk codes and observed spectra (3700 to 10,500 \r{A}) from the ESPaDOnS instrument on CFHT. The photoionizing radiation of the central star is assumed to be the sole source of input energy for the disk. We model optical and near-infrared emission lines that are thought to form in this region using standard techniques that have been successful in modeling the spectra of Classical Be stars. By comparing synthetic line profiles of hydrogen, helium, iron and calcium with the observed line profiles, we try to constrain the geometry, density structure, and kinematics of the gaseous disk. Reasonable matches have been found for all line profiles individually; however, no disk density model based on a single power-law for the equatorial density was able to simultaneously fit all of the observed emission lines. Amongst the emission lines, the metal lines, especially the Ca II IR triplet, seem to require higher disk densities than the other lines. Excluding the Ca II lines, a model in which the equatorial disk density falls as $10^{-10} (R_{*}/R)^3 g\,cm^{-3}$ seen at an inclination of 45{\deg} for a $50\,R_{*}$ disk provides reasonable matches to the overall line shapes and strengths. The Ca II lines seem to require a shallower drop off as $10^{-10} (R_{*}/R)^2 g\,cm^{-3}$ to match their strength. More complex disk density models are likely required to refine the match to the BD+65 1637 spectrum.

Non-LTE calculations are performed for NII in stellar atmospheric models appropriate to main sequence B-stars to produce new grids of equivalent widths for the strongest NII lines commonly used for abundance analysis. There is reasonable agreement between our calculations and previous results, although we find weaker non-LTE effects in the strongest optical NII transition. We also present a detailed estimation of the uncertainties in the equivalent widths due to inaccuracies in the atomic data via Monte Carlo simulation and investigate the completeness of our model atom in terms of included energy levels. Uncertainties in the basic NII atomic data limit the accuracy of abundance determinations to ~+/-0.10 dex at the peak of the NII optical spectrum near Teff~ 24,000 K., Comment: 18 pages; 11 figures

Omicron Aquarii is late-type, Be shell star with a stable and nearly symmetric H$\alpha$ emission line. We combine H$\alpha$ interferometric observations obtained with the Navy Precision Optical Interferometer (NPOI) covering 2007 through 2014 with H$\alpha$ spectroscopic observations over the same period and a 2008 observation of the system's near-infrared spectral energy distribution to constrain the properties of $o$~Aqr's circumstellar disk. All observations are consistent with a circumstellar disk seen at an inclination of $75\pm\,3^{\circ}$ with a position angle on the sky of $110\pm\,8^{\circ}$ measured E from N. From the best-fit disk density model, we find that 90\% of the H$\alpha$ emission arises from within $9.5$ stellar radii, and the mass associated with this H$\alpha$ disk is $\sim 1.8\times10^{-10}$ of the stellar mass and the associated angular momentum, assuming Keplerian rotation for the disk, is $\sim 1.6\times10^{-8}$ of the total stellar angular momentum. The occurrence of a central quasi-emission (CQE) feature in Mg\,{\sc ii} $\lambda\,4481$ is also predicted by this best-fit disk model and the computed profile compares successfully with observations from 1999. To obtain consistency between the H$\alpha$ line profile modelling and the other constraints, it was necessary in the profile fitting to weight the line core (emission peaks and central depression) more heavily than the line wings, which were not well reproduced by our models. This may reflect the limitation of assuming a single power-law for the disk's equatorial density variation. The best-fit disk density model for $o$~Aqr predicts that H$\alpha$ is near its maximum strength as a function of disk density, and hence the H$\alpha$ equivalent width and line profile change only weakly in response to large (factor of $\sim 5$) changes in the disk density., Comment: 16 pages; 16 figures