OHANA is an interferometric snapshot survey of the gaseous circumstellar environments of hot stars, carried out by the VLTI group at the Paranal observatory. It aims to characterize the mass-loss dynamics (winds/disks) at unexplored spatial scales for many stars. The survey employs the unique combination of AMBER's high spectral resolution with the unmatched spatial resolution provided by the VLTI. Because of the spatially unresolved central OBA-type star, with roughly neutral colour terms, their gaseous environments are among the easiest objects to be observed with AMBER, yet the extent and kinematics of the line emission regions are of high astrophysical interest.

Spectropolarimetry is a powerful tool used to probe fundamental properties of stars that cannot typically be measured in any other way. A new generation of high-resolution spectropolarimeters (ESPaDOnS at the Canada-France-Hawaii telescope, Narval at the Télescope Bernad Lyot, and HARPSpol at the 3.6-m ESO telescope) and dedicated observing campaigns (such as the Magnetism in Massive Stars (MiMeS) project) have led to significant improvements in both our observational and theoretical understanding of the underlying physics governing massive stars. In this article I review recent advances in the field of stellar magnetism of massive stars acquired using spectropolarimetry.

Many astronomical sources of radiation emit polarised radiation, for example because of the presence of a disk which produces linear polarisation by scattering some photospheric radiation, or because of the presence of a magnetic field, which leads to circular and sometimes linear polarisation of spectral line profiles. Measuring the wavelength dependence of the polarisation of radiation from such sources can reveal valuable and interesting constraints on the nature of the objects observed. This paper summarises the basic ideas of spectropolarimetry and describes some of the information it can provide.

As per the recent study by the MiMeS collaboration, only about 10% of massive stars possess organized global magnetic fields, typically dipolar in nature. The competition between such magnetic fields and highly non-linear radiative forces that drive the stellar winds leads to a highly complex interaction. Such an interplay can lead to a number of observable phenomena, e.g. X-ray, wind confinement, rapid stellar spindown. However, due to its complexity, such an interaction cannot usually be modeled analytically, instead numerical modeling becomes a necessary tool. In this talk, I will discuss how numerical magnetohydrodynamic (MHD) simulations are employed to understand the nature of such magnetized massive star winds.

It is now well established that a fraction of the massive (M > 8 M⊙) star population hosts strong, organised magnetic fields, most likely of fossil origin. The details of the generation and evolution of these fields are still poorly understood. The BinaMIcS project takes an important step towards the understanding of the interplay between binarity and magnetism during the stellar formation and evolution, and in particular the genesis of fossil fields, by studying the magnetic properties of close binary systems. The components of such systems are most likely formed together, at the same time and in the same environment, and can therefore help us to disentangle the role of initial conditions on the magnetic properties of the massive stars from other competing effects such as age or rotation. We present here the main scientific objectives of the BinaMIcS project, as well as preliminary results from the first year of observations from the associated ESPaDOnS and Narval spectropolarimetric surveys.

The majority of massive stars are members of binary systems. However, in order to understand their evolutionary pathways, mass and angular momentum loss from these systems needs to be well characterized. Self-consistent explanations for their behavior across many wavelength regimes need to be valid in order to illuminate key evolutionary phases. I present the results of linear spectropolarimetric studies of three key binaries (β Lyrae, V356 Sgr, V444 Cyg, and WR 140) which reveal important geometric information about their circumstellar material. β Lyrae exhibits a repeatable discrepancy between secondary eclipse in the total and polarized light curves that indicates an accretion hot spot has formed on the edge of the disk in the system. The existence of this hot spot and its relationship to bipolar outflows within the system is important in the understanding of mass transfer dynamics in Roche-lobe overflow binaries. Preliminary work on V356 Sgr suggests the system maybe surrounded by a common envelope. V444 Cyg shows evidence that its shock creates a cone with a large opening angle of missing material around the WN star. This suggests the effects of radiative inhibition or braking, can be significant contributors to the location and shape of the shock within colliding wind binaries. The intrinsic polarization component of WR 140 is likely due to the formation of dust within the system near periastron passages. Continued work on these and additional objects will provide new and important constraints on the mass loss structures within binary systems.

The B fields in OB stars (BOB) survey is an ESO large programme collecting spectropolarimetric observations for a large number of early-type stars in order to study the occurrence rate, properties, and ultimately the origin of magnetic fields in massive stars. As of July 2014, a total of 98 objects were observed over 20 nights with FORS2 and HARPSpol. Our preliminary results indicate that the fraction of magnetic OB stars with an organised, detectable field is low. This conclusion, now independently reached by two different surveys, has profound implications for any theoretical model attempting to explain the field formation in these objects. We discuss in this contribution some important issues addressed by our observations (e.g., the lower bound of the field strength) and the discovery of some remarkable objects.

σ Ori E (HD 37479) is the prototypical helium-strong star shown to harbor a strong magnetic field, as well as a magnetosphere consisting of two clouds of plasma. The observed optical (ubvy) light curve of σ Ori E is dominated by eclipse features due to circumstellar material, however, there remain additional features unexplained by the Rigidly Rotating Magnetosphere (RRM) model of Townsend & Owocki (2005). Using the technique of magnetic Doppler imaging (MDI), spectropolarimetric observations of σ Ori E are used to produce maps of both the magnetic field topology and various elemental abundance distributions. We also present an analysis utilizing these computed MDI maps in conjunction with non-local thermodynamical equilibrium TLUSTY models to study the optical brightness variability of this star arising from surface inhomogeneities. It has been suggested that this physical phenomena may be responsible for the light curve inconsistencies between the model and observations.

A quite remarkable aspect of non-interacting O-stars with detected surface magnetic fields is that they all are very slow rotators. This paper uses this unique property to first demonstrate that the projected rotational speeds of massive, hot stars, as derived using current standard spectroscopic techniques, can be severely overestimated when significant “macroturbulent” line-broadening is present. This may, for example, have consequences for deriving the statistical distribution of rotation rates in massive-star populations. It is next shown how such macroturbulence (seemingly a universal feature of hot, massive stars) is present in all but one of the magnetic O-stars, namely NGC 1624-2. Assuming then a simple model in which NGC 1624-2's exceptionally strong, large-scale magnetic field suppresses atmospheric motions down to layers where the magnetic and gas pressures are comparable, first empirical constraints on the formation depth of this enigmatic hot-star macroturbulence is derived. The results suggest it originates in the thin sub-surface convection zone of massive stars, consistent with a physical origin due to, e.g., stellar pulsations excited by the convective motions.

Various theories have been proposed to predict how mass loss depends on the stellar rotation rate, both in terms of its strength, as well as its latitudinal dependence, crucial for our understanding of angular momentum evolution. Here we discuss the tool of linear spectropolarimetry that can probe the difference between mass loss from the pole versus the equator. Our results involve several groups of O stars and Wolf-Rayet stars, involving Oe stars, Of?p stars, Onfp stars, as well as the best candidate gamma-ray burst progenitors identified to date.

The stars of the middle main-sequence have relatively quiescent outer layers, and unusual chemical abundance patterns may develop in their atmospheres, revealing the action of such subsurface phenomena as gravitational settling and radiatively driven levitation of trace elements, and their competition with mixing processes such as turbulent diffusion. We report the discovery of the time evolution of such chemical tracers through the main-sequence lifetime of magnetic chemically peculiar stars.

Magnetic fields play a significant role in the evolution of massive stars. About 7% of massive stars are found to be magnetic at a level detectable with current instrumentation (Wade et al. 2013) and only a few magnetic O stars are known. Detecting magnetic field in O stars is particularly challenging because they only have few, often broad, lines to measure the field, which leads to a deficit in the knowledge of the basic magnetic properties of O stars. We present new spectropolarimetric Narval observations of ζ Ori A. We also provide a new analysis of both the new and older data taking binarity into account. The aim of this study was to confirm the presence of a magnetic field in ζ Ori A. We identify that it belongs to ζ Ori Aa and characterize it.

Cool supergiants offer a good opportunity to study the interplay of magnetic fields and stellar evolution. We present the results of spectropolarimetric study of the cool supergiants and classical Cepheids η Aql and ζ Gem.

For 30 years, cyclical wind variability in OB stars has puzzled the astronomical community. Phenomenological models involving co-rotating bright spots provide a potential explanation for the observed variations, but the underlying physics remains unknown. We present recent results from hydrodynamical simulations constraining bright spot properties and compare them to what can be inferred from space-based photometry. We also explore the possibility that these spots are caused by magnetic fields and discuss the detectability of such fields.

The MiMeS project demonstrated that a small fraction of massive stars (around 7%) presents large-scale, stable, generally dipolar magnetic fields at their surface. These fields that do not present any evident correlations with stellar mass or rotation are supposed to be fossil remnants of the initial phases of stellar evolution. They result from the relaxation to MHD equilibrium states, during the formation of stable radiation zones, of initial fields resulting from a previous convective phase. In this work, we present new theoretical results, where we generalize previous studies by taking rotation into account. The properties of relaxed fossil fields are compared to those obtained when rotation is ignored. Consequences for magnetic fields in the radiative envelope of rotating early-type stars and their stability are finally discussed.

It is shown that the diagnostics from an observed circularly polarized line in a rapidly rotating star are directly interpretable, not in terms of the observed Stokes V profiles, but in terms of its antiderivative with respect to wavelength (in velocity units if preferred). This also leads to a new mean-field diagnostic that is just as easily obtained as the standard “center of gravity” approach, and is less susceptible to cancellation if the line-of-sight field changes sign over the face of the star.

The intrinsic linearly polarized light arising from electron scattering of stellar radiation in a non-spherically symmetric distribution of gas is a characterizing feature of classical Be stars. The distinct polarimetric signature provides a mean for directly probing the physical and geometric properties of the gaseous material enveloping these rapidly-rotating massive stars. Using a Monte Carlo radiative transfer computation and a self-consistent radiative equilibrium solution for the circumstellar gas, we explore the role of this observable signature in investigating the dynamical nature of classical Be star disks. In particular, we focus on the potential for using linearly polarized light to develop diagnostics of mass-loss events and to trace the evolution of the gas in a circumstellar disk. An informed context for interpreting the observed linear polarization signature can play an important role in identifying the physical process(es) which govern the formation and dissipation of the gaseous disks surrounding classical Be stars.

Many OB stars show unexplained cyclical variability in their winds and in many optical lines, which are formed at the base of the wind. For these stars no dipolar magnetic fields have been detected. We propose that these cyclical variations are caused by the presence of multiple, transient, short-lived, corotating magnetic loops, which we call “stellar prominences”. We present a simplified model representing these prominences as corotating spherical blobs and fit the rapid variability in the Heii λ4686 line of the O supergiant λ Cep for time-resolved spectra obtained in 1989. Our conclusions are: (1) From model fits we find that the life time of the prominences varies, and is between 2–7 h. (2) The adopted inclination angle is 68° with a rotation period of ≈ 4.1 d (but not well constrained). (3) The contribution of non-radial pulsations is negligible (4) Similar behavior is observed in at least 4 other O stars. We propose that prominences are a common phenomenon among O stars.

We present the first results for the estimation of gravity and effective temperature of poorly studied chemically peculiar stars recently observed with the spectropolarimeter ESPaDOnS at CFHT in the frame of the VeSElkA (Vertical Stratification of Elements Abundance) project. A grid of theoretical stellar atmosphere models with the corresponding fluxes has been calculated using the PHOENIX code. We have used these fluxes to fit Balmer line profiles employing the code FITSB2 that produces estimates of the effective temperature, surface gravity and radial velocity for each star.

The new spectropolarimetric observations of HD 22920 with ESPaDOnS at CFHT reveal a strong variability of its spectral line profiles with the phase of stellar rotation. We have obtained Teff = 13640 K, logg=3.72 for this star from the best fit of its nine Balmer line profiles. The respective model of stellar atmosphere was calculated to perform abundance analysis of HD 22920 using the spectra obtained for three different phases of stellar rotation. We have found that silicon and chromium abundances appear to be vertically stratified in the atmosphere of HD 22920. Meanwhile, silicon shows hints for a possible variability of vertical abundance stratification with rotational phase.