Caltech
Hot And Cool:
Bridging Gaps in Massive Star Evolution
NASA Herschel Science Center





Program

Download the full workshop schedule (pdf)

Posters

Ronny BlommeRoyal Observatory BelgiumThe colliding winds of Cyg OB2 No. 8A
Howard BondSTScIThe NGC 300 Transient -- Supernova or Giant Eruption?
Maria DroutUniversity of IowaFilling the Yellow Void: A Census of F & G supergiants in M31
Ariane LanconObservatoire de StrasbourgPHOENIX spectra of red supergiants --- Varying surface abundances and microturbulence
Pat MorrisNHSC/CaltechRevealing "Hidden" Hot Massive Star Populations in the Galaxy
Jesus Toala SanzCentro de Radioastronomia y Astrofisica, UNAMRadiation-hydrodynamic models of the evolving circumstellar medium around massive stars
Stefanie WachterSSC/CaltechMid-Infrared Circumstellar Shell Sources Discovered with Spitzer: An Obscured Population of Massive Stars?

 

Talks

Philip BennettSaint MaryEmpirical Constraints on Modeling RSG Winds
Alceste BonanosSTScIA Survey of the Most Massive Stars in the Local Universe
Matteo CantielloAstronomical Institute UtrechtConvection zones in the envelope of hot, massive stars
Sabina ChitaAstronomical Institute UtrechtMultiple ring nebulae around bluesupergiants
Isabelle CherchneffInstitute for Astronomy, ETH ZurichDust formation in massive stars and their explosive ends
Cian CrowleyTrinity CollegeWinds from Cool Evolved Stars
Ben DaviesUniversity of LeedsYoung Massive Clusters as probes of massive stellar evolution
Nancy Elias-RosaSSC/CaltechThe Progenitors of Recent Core-Collapse Supernovae
Jose GrohMax-Planck-Institut fuer RadioastronomieHot and cool: on the nature and evolution of stellar and wind parameters of the prototype LBV AG Carinae across the HR diagram
Graham HarperUniversity of ColoradoWhy Multi-Wavelength Studies of Inhomogeneous Atmospheres are Essential: Betelgeuse - A Case Study.
Alexander HegerUniversity of MinnesotaThe Final Evolution Stages of Massive Stars
D. John HillierUniversity of PittsburghBasic Parameters of Hot Massive Stars
Raphael HirschiKeele UniversityStellar Evolution in the upper HRD
Roberta HumphreysUniversity Of MinnesotaThe Circumstellar Environments of the Cool Hypergiants: Post-RSG Evolution and Blue-Loops
Ciska KemperUniversity of ManchesterMineralogy as an evolutionary clock
Laszlo KissUniversity of SydneyCHARA/MIRC interferometry of red supergiant stars: hotspots, diameters and effective temperatures
Gloria KoenigsbergerUniversidad Nacional Autonoma de Mexico-ICFAsynchronous rotation in binaries
Ariane LanconObservatoire de StrasbourgConstraints on red supergiant evolution from the integrated light of star clusters.
Norbert LangerUniversiteit UtrechtThe stars in between: blue supergiants and the circumstellar medium
William LatterNHSC/CaltechOpportunties with the Herschel Space Observatory
Danny LennonESA/STScIProperties of B-type supergiants in the Milky Way, LMC and SMC
Claus LeithererSTScIStellar Mass Loss in the Upper HRD
Emily LevesqueU. HawaiiThe Physical Properties of Red Supergiants
Alex LobelRoyal Observatory of BelgiumRadiative Transfer Modeling the Winds and Circumstellar Environments of Hot and Cool Massive Stars
Douglas LeonardSan Diego State UniversitySeeking Supernova Progenitors in Pre-Explosion Images
Philip MasseyLowell ObservatoryA Census of Massive Stars Across the HRD: What We Know and What We Don't
Jon MauerhanIPAC/CaltechWolf-Rayet/O Binaries Near the Galactic Center
Maryam ModjazUC Berkeley/Miller Institute Supernova 2008D/X-ray Transient 080109: Death of a Stripped-Envelope Massive Star
John MonnierUniversity of MichiganInfrared Interferometry of the Upper HRD
Maria-Fernanda NievaMax-Planck-Institute for AstrophysicsHigh-precision determinations of stellar parameters & chemical abundances in early B-type stars
Stanley OwockiUniiversity of DelawareHot-Star Mass Loss Mechanisms: Winds and Outbursts
Bertrand PlezGRAAL, Univ. MontpellierModel atmospheres for cool massive stars
Jose L PrietoOhio State UniversityA New Class of Luminous Transients and a first Census of their Progenitors
Norbert PrzybillaRemeis Observatory BambergTepid Supergiants: Chemical Signatures of Stellar Evolution & the Extent of Blue Loops
Nathan SmithUC BerkeleyMass Loss and Pre-SN evolution of Massive Stars
Francesca ValsecchiNorthwestern UniversityM33 X-7, The First Eclipsing Black Hole X-Ray Binary.
Nolan WalbornSTScIAge Paradigms for Massive Young Clusters
Schuyler Van DykSSC/CaltechExplosions of LBV and Post-LBV Stars
Jacco van LoonKeele UniversityThe effects of red supergiant mass loss on supernova ejecta and the circumburst medium
Rens WatersUniversity of AmsterdamCircumstellar material in massive stars
Sung-Chul YoonUCO/Lick, UCSCProgenitor stars of Type Ib/c supernovae in close binary systems

 

Poster Abstracts

Cyg OB2 No. 8A is an O6 If + O5.5 III(f) colliding-wind binary with a period of 21.9 days. At the shocks in the colliding-wind region, some of the electrons get Fermi-accelerated to relativistic speeds and emit synchrotron radiation as they spiral in the magnetic field. Our Very Large Array (VLA) observations have detected this non-thermal radiation. The radio fluxes vary in a correlated way with orbital phase. This finding is actually quite unexpected, as all the synchrotron emission is predicted to be absorbed by the free-free absorption in the stellar wind material of both binary components. We develop radiative transfer models for this system, which confirm this prediction. In order to avoid the discrepancy between theory and observations, we propose that the stellar wind material is not smoothly distributed, but is clumped and porous to radiation. Porous winds have previously been proposed to solve a number of problems in single-star stellar winds. Our models show that, with a sufficiently large porosity, some of the synchrotron radiation can indeed escape through the stellar winds.

A new 14th magnitude star was discovered by Monard in the Sculptor Group spiral galaxy NGC 300 in April 2008. At the galaxy's distance of ~1.9 Mpc, the visual absolute magnitude at maximum was about -12, intermediate in brightness between classical novae and Type II core-collapse supernovae. HST/WFPC2 images in June and September 2008, allow us to locate the outburst site precisely in archival pre-outburst HST/ACS images. The site lies in an area of active star formation, but there is no optical progenitor star to a limit of ~28 mag. Archival Spitzer images from 2003 and late 2007 show a luminous infrared star at the site detected at 3.6 through 24 microns yielding a luminosity of 9 X 104 L/Lsun for the progenitor. Photometric monitoring with the SMARTS 1.3m telescope at CTIO shows a smoothly declining optical light curve, with a suggestion of a flattening in the latest observations. Spectroscopic monitoring with the SMARTS 1.5m shows an underlying continuum resembling an F-type supergiant with strong emission lines of the Balmer series, Ca II, and [Ca II]. Halpha and Ca II emission lines have shown a double structure, suggesting a bipolar outflow from the central object. We interpret the event as an outburst of a heavily dust-obscured OH/IR source currently on a blue-loop to warmer temperatures. SN 2008S may be another member of this class.

The F and G supergiant content of nearby galaxies provide a critical test of stellar evolution theory, ``bridging the gap" between the hot, massive stars and the cool red supergiants. But, this region of the color-magnitude diagram is dominated by foreground contamination. Fortunately, the large negative systemic velocity of M31, coupled to its high rotation rate, provides the meansfor separating the foreground dwarfs from the bona fide supergiants within M31. During the fall of 2006 we obtained radial velocities of 2901 individual targets within the correct color-magnitude range using the 6.5-m MMT telescope and Hectospec fiber spectrograph.

After a comparison with the velocities predicted by the rotation curve of M31, we found 54 definite and 66 likley M31 yellow supergiants, indicating a foreground contamination > 97%. We compare the location of these supergiants to the Geneva evolutionary tracks in the HR-diagram. There is some indication that the current tracks may not extend to sufficiently cool temperatures at high luminosities, but that result rests on three "probable" members. A much stronger conclusion, however, can be drawn from the relative number of F and G supergiants as a function of luminosity (mass): the current tracks predict many times more high luminosity F and G supergiants that are actually seen, assuming a normal IMF, suggesting that their lifetimes are too long (relative to that of lower-mass supergiants) by about a factor of 100.

No abstract provided.

Efforts to detect new Wolf-Rayet (WR) stars in relation to massive OB star and Red Supergiant populations in the Milky Way over the last decade have been motivated by testing stellar evolution model predictions on lifetimes in different core burning phases and to better establish their energetic contributions in galaxies. A large number of WR stars, up to 90% of the predicted population, have been suspected to be hidden behind large amounts of extinction, more than 10 magnitues in the K band, especially in dense stellar clusters of Galactic HII regions and especially towards the Galactic Center. Recent near-IR photometric and spectroscopic surveys have in fact revealed several concentrations of WR stars. A near-IR 2MASS-based survey to a completeness of ~12mag in K however is hampered by some degeneracies in colors with transition OB emission line stars and dusty young stars. We recognize that the unique free-free continuum and H/He/C emission line properties which explain their color behavior can also be used to remove this degeneracy to increasing degrees with the inclusion of longer wavelength colors. We demonstrate how combined near- to mid-IR colors set a candidate population of WR stars apart from other object classes, and give the status of new discoveries of very reddened stars in the "open" field versus clusters/associations and discuss the impacts implied on massive star population balance in the Galaxy.

We present the results of time-dependent, radiation-hydrodynamic models of the evolving circumstellar medium around stars with initial masses of 40 and 60 solar masses. Our models include photoionization by the central star, together with its stellar wind, from the main sequence through the red supergiant or LBV intense mass-loss stages, to the final fast-wind Wolf-Rayet stage. The evolving mass-loss rates are taken from published models both with and without stellar rotation, with the stellar wind velocity and ionizing photon rate being calculated self-consistently. Our models show the difference that the assumption of a rotating star can make to the resulting circumstellar medium, which is principally due to the difference in length of the intense mass-loss stages. We model the red supergiant and Wolf-Rayet stages in 2-D axisymmetry in order to highlight the formation of structure due to instabilities in the cold shell-fast wind interaction region. Finally, we include thermal conduction in some of our models to investigate whether evaporated material could be responsible for the soft X-ray emission seen in XMM-Newton observations of the Wolf-Rayet wind bubble S308.

We have discovered a large number of symmetric shells around luminous central sources at 24 microns with MIPS on-board the Spitzer Space Telescope. Most of these shells are not visible in the shorter wavelengths bands of IRAC or archival 2MASS and optical images. A careful archival follow-up effort has revealed 99% of these sources to be previously unknown. We suspect that a fraction of the central stars represent a population of highly obscured massive stars. We will present early results from follow-up observations aimed at characterizing this population.

 

Talk Abstracts

Stars between about 4 and 25 solar masses spend a significant fraction of their post-main sequence lifetime as red supergiants (RSGs) and lose material via stellar winds during this period. For RSGs more massive than 10 solar masses, this mass loss becomes of evolutionary significance, and probably determines the upper mass limit of RSGs in the H-R diagram. Despite decades of observations, the driving mechanism responsible for mass loss in RSGs remains unknown. Mainly this is because the optical spectrum accessible from the ground provides almost no useful wind diagnostics, and what information is obtained is spatially averaged over the entire wind volume. However, within the last decade, Hubble Space Telescope (HST) observations of many useful ultraviolet wind diagnostics have been obtained at high S/N and spectral resolution. In particular, an extensive series of HST observations of RSGs in eclipsing binaries used their UV-bright companions to probe circumstellar environments and provide spatially resolved observations of RSG winds. By observing RSG binaries near and during eclipse, the structure of the supergiants' outer atmospheres can be mapped, and empirical wind models constructed. In this talk, I will review the historical use of binaries to probe RSG winds, and present some empirical models and constraints on RSG wind properties.

Despite the large impact very massive stars (>30 Mo) have in astrophysics, their fundamental parameters remain uncertain. The most accurate method for deriving masses, radii and luminosities of such distant stars is to measure them in eclipsing binary systems. Currently, the most massive eclipsing binary accurately measured is WR20a, which consists of two 80 solar mass stars in a 3.7 day orbit. In total, only ~15 very massive stars (>30 Mo) belonging to our Galaxy and Local Group galaxies have accurate determinations of their parameters. I will present the first results of a wide-ranging survey targeting the brightest and thus most massive stars in eclipsing binaries in both young massive clusters in the Milky Way and in nearby galaxies. The measurement of fundamental parameters for massive stars at a range of metallicities will provide much needed constraints on theories that model the formation and evolution of massive stars and will observationally probe the upper limit on the stellar mass.

We present results from the first extensive study of convection zones in the envelopes of hot massive stars. These regions are caused by opacity peaks associated with iron and helium ionization. Such convective regions can be very close to the stellar surface.

Recent observations of microturbulence in massive stars from the VLT-Flames survey are in very good agreement with our predictions concerning the occurrence and the strength of sub-surface convection in hot stars. We argue that convection close to the surface may affect the stellar mass loss by triggering wind clumping.

In the course of the life of a massive star, wind-wind interaction can give rise to the formation of circumstellar nebulae which are both predicted and observed in the nature. We present generic model calculations to predict the properties of such nebulae for blue supergiants. From stellar evolution calculations including rotation, we obtain the time dependence of the stellar wind properties and of the stellar radiation field. These are used as input for hydro-calculations of the circumstellar medium throughout the star's life. Here, we present the results for a rapidly rotating 12 solar masses single star. This star undergoes a blue loop during its post main sequence evolution, at the onset of which its contraction spins it up close to critical rotation. Due to the consequent anisotropic mass loss, the blue supergiant wind sweeps up the preceding slow wind into an hour glass structure. Its collision with the previously formed spherical red supergiant wind shell forms a short-lived luminous nebula consisting of two polar caps and a central inner ring. With time, the polar caps evolve into mid-latitude rings which gradually move toward the equatorial plane while the central ring is fading. These structures are reminiscent to the observed nebulae around the blue supergiant Sher 25 and the progenitor of SN 1987A. The simple model of an hour glass colliding with a spherical shell retrieves most of the intriguing nebula geometries discovered around blue supergiants, and suggests them to form an evolutionary sequence. Our results indicate that binarity is not required to obtain them.

Massive stars in the late stages of their evolution are well known dust makers. Of particular interest are Wolf-Rayet stars and the explosive end of Red Supergiants as Type II Supernovae. Despite numerous observational evidence of dust formation in the massive winds of carbon-rich Wolf-Rayet (WC) stars, the mechanisms for solid condensation in those fast, helium-poor outflows are far from been understood. They probably involve highly clumpy winds and/or the presence of a stellar companion. Type II Supernovae are too believed to be important dust providers to the interstellar medium. In the early universe, supernovae explosion of very massive stellar progenitors may be the only dense locus of enriched material in which dust condenses. Primordial SNe should thus be crucial contributors to the dust budget at high redshifts. However, there is no observational evidence for large amounts of dust formed by local core-collapse SNe. I shall review the dust formation processes at play in WC and Type II supernovae and discuss their relevance to the dust budget in the far and near universe.

The origin and characterization of winds from non-dusty evolved K and M stars remains an outstanding issue. We are currently carrying out a program to diagnose the inner wind region of a small sample of giant stars using eclipse mapping of symbiotic binaries. I shall relate our results to those from similar wind studies of supergiants, and also discuss the applicability of these results to isolated stars. In addition, I will examine the current status of the field, and look at some the most promising options for future progress.

Young Massive Clusters (YMCs) represent ideal testbeds in which to study massive stellar evolution as they contain large, coeval, chemically homogeneous, samples of massive stars. By studying YMCs with a range of ages (and hence turn-off masses), we can investigate the post main-sequence evolution of massive stars as a function of initial mass. Recent discoveries of YMCs over a range of ages within our own Galaxy - where we can successfully resolve individual stars - offers the unprecedented opportunity to test our ideas of massive stellar evolution. Here, I review the many recent works in this field, how YMCs both reaffirm our understanding of stellar evolution in some areas, and force us to return to the drawing board in others.

We will present the results of our analysis of Hubble Space Telescope (HST) and deep ground-based images, to isolate the massive progenitor stars of several recent core-collapse supernovae (SNe). The identification of the progenitors is facilitated in a few cases by high-precision astrometry based on our HST ToO imaging of the SNe at late times. We will also intend to present the trends of core-collapse SN sites as a function of their local stellar environments, based on HST archival images. We will discuss the implications of these results on the nature of the SN progenitors.

The Luminous Blue Variable (LBV) phase is characterized by strong photometric and spectroscopic variability on timescales from days to decades, arising from changes in stellar and wind parameters. Key changes have been incorporated into radiative transfer codes during the last decade, such as the inclusion of wind clumping and a consistent treatment of line blanketing. In this talk I will present how the determination and understanding of the properties of LBVs are changed when those effects are taken into account. I will discuss the results obtained for the prototype LBV AG Carinae, which has been the most variable object among the Galactic LBVs during the last decades. We performed a detailed spectroscopic analysis of AG Car based on UV, optical, and near-IR spectra taken during the last 20 years. The radiative transfer code CMFGEN was used to obtain the temporal evolution of the mass-loss rate, wind terminal velocity, effective temperature, stellar radius, luminosity, and rotational velocity. We obtained that Teff varies in the range 8kK-23kK, while the bolometric luminosity is not constant as the star crosses the HR-diagram: L clearly decreases from minimum to maximum. We determined that AG Car was at different sides of the bi-stability limit in different epochs, but there is no clear evidence of a jump in the mass-loss rate as a function of the effective temperature, as suggested by previous works. When allowing for time-dependent effects, we determined that the mass-loss rate reaches roughly a constant value during maximum, suggesting that the "eruption" during the S-Dor cyles is actually an increase in the mass-loss rate by a factor of ~4, which begin years before the maximum in the lightcurve. The high rotational velocity of AG Car detected during minimum suggests that M > 70 solar masses, which is significantly higher than previous estimates, and put strong constraints on the progenitor, current evolutionary stage, and fate of LBVs.

Despite being one of the most studied stars, the massive M supergiant Betelgeuse had us fooled. An unfortunate convergence of observational interpretation, semi-empirical analyses, and theoretical modeling led to the notion that the accelerating wind was essentially at chromospheric temperatures. Only with high spatial-resolution multi-wavelength datasets covering a wide swath of the electromagnetic spectrum did it finally become apparentthat the bulk of the wind is much cooler, while encompassing tiny volumes of hot UV emitting plasma. New high spectral-resolution TEXES observations of mid-IR [Fe II] emission are now revealing the dynamical nature of the pervasive cool plasma. Extended and inhomogeneous atmospheres appear to be a common property of early M supergiants.

I will discuss the evolution of massive stars through their final stages before their explosive death. The evolution in different mass regimes from initial masses of 10 solar masses up to the most massive stars will be shown.

With recent progress in atmospheric modeling and the availability of extensive spectroscopic observations, there have been significant advancements in our knowledge of the fundamental properties of massive stars. We highlight recent results for the galaxy, and Magellanic clouds, with particular emphasis on the effective temperature-scale, abundances, and mass-loss rates. Deficiencies in current spectroscopic analyses will be highlighted as will the remaining conflicts between spectroscopic analyses and evolutionary calculations.

I will present stellar evolution models of massive stars from the ZAMS up to the advanced stages at various metallicities. I will highlight the impact of mass loss and its metallicity dependence on the evolution of the most massive stars. Finally, I will describe the evolutionary tracks in the HR diagram and compare the models to observations.

The evolved yellow and red hypergiants near the empirical upper luminosity boundary all show evidence for instabilities and high mass loss but a few are distinguished by extensive circumstellar ejecta and evidence for episodic mass loss events and eruptions. What distinguishes these stars and what are the implications for their evolutionary state? I will also present some new results for lower mass massive stars.

An extensive literature detailing the dust mineralogy around massive stars is available, showing a large variety of dust properties and compositions across the HR-diagram. In this review talk, I will discuss the formation and processing mechanisms of dust, and aim to correlate the dust characteristics with the physical properties and thus the evolutionary status of the stars. So far, most of the work done has been on Galactic sources, but recent studies performed with Spitzer have extended parameter space to the lower metallicity environments of the LMC and SMC, and I will also highlight a few of those results.

We have obtained H-band interferometric observations of three galactic red supergiants stars using the MIRC instrument on the CHARA array. The targets include AZ Cyg, a field RSG and two members of the Perseus Double Cluster, RS Per and T Per. From fitting visibilities and closure phases we derive surface maps of the stars with a range of algorithms. We find evidence of departures from circular symmetry in each cases, which can be modelled with the presence of hotspots. This is the first detection of these features in the H-band, which allowed us mapping surface structures as close to the photosphere as possible. The measured mean diameteres and the spectral energy distributions were combined to determine effective temperatures for the sample. The results give further support to the recently derived hotter temperature scale of red supergiant stars, which has been evoked to reconcile the empirically determined physical parameters and stellar evolutionary theories. We see a possible correlation between spottednes and mid-IR emission of the circumstellar dust, suggesting a connection between mass-loss and the mechanism that generates the spots. Finally, the two cluster member RSGs are located at a distance known to better than 5%, implying that they could play an important role in the absolute calibration of red supergiant stars.

Asynchronous rotation in binary stars produces perturbations on the stellar surface that are not generally fully appreciated. In this paper we will describe how asynchronous rotation may impact the determination of basic parameters, such as stellar mass, by its effects on the shape of photospheric absorption lines. In addition, we will discuss its potential role in giving rise to phenomena involving asymmetric mass-loss and X-ray emission. We propose that as stars evolve off the Main Sequence and their expanding radii lead to ever increasing tidal forcing, mass-loss may be enhanced through viscous dissipation of tidal energy. In stars with luminosities approaching the Eddington luminosity, the additional energy contribution may lead to episodic mass ejections or even large eruptions, once a critical radius is attained. Because the perturbations produced by tidal effects are not equally distributed over the stellar surface, the associated enhanced mass-loss is expected to depart significantly from spherical symmetry.

In studies of red supergiants, empirical samples of stars with known cluster memberships are obviously useful, as the cluster environment provides extra information on age and metallicity. Can more distant clusters in star forming galaxies also help us constrain the physics of red supergiants, even though they are not resolved into stars? Certainly, but the task of extracting these constraints is difficult. This talk will address some of the relevant issues (e.g. spectral libraries for population synthesis, degeneracies, stochasticity) and attempt to highlight some promising paths and some serious limitations.

No abstract provided.

A general description of the far-infrared/submillimeter Herschel Space Observatory mission will be provided with emphasis on coming opportunities for observations by the astronomical community. Large, Key Programs have been accepted for early execution on the Observatory. A brief discussion of what observations are planned relating to the general topic of this workshop will be discussed.

B-type supergiants have always been problematic for our understanding of massive stars; they lie in the theoretical post main-sequence gap and as such should not exist according to current theories; the progenitor of SN1987A in the LMC was a mid-B supergiant which to this day causes considerable discussion; their derived mass-loss rates too are controversial as they disagree with the predicted trend with effective temperature; and there is now evidence that their surface abundaces do not solely reflect the impact of rotational mixing on the main-sequence. In this paper we review the properties of B-type supergiants in the Milky Way, SMC and LMC, presenting new results for the previously neglected LMC supergiants, and discuss their chemical compositions, rotational velocities and mass-loss rates. We attempt to put these results into the context of massive star evolution across the HRD at different metallicities, and look at relationships with hotter and cooler massive stars.

I will discuss empirically derived mass-loss rates of massive stars in the cool and hot portions of the Hertzsprung-Russell diagram. The stellar types covered are red supergiants, AGB stars, luminous blue variables, O stars, and Wolf-Rayet stars. Comparisons with theoretical predictions will be made.

In this review talk, I will summarize what we have learned about the nature of stars that ultimately explode as core-collapse supernovae from the examination of images taken prior to the explosion. By registering pre-supernova and post-supernova images, usually taken at high resolution using either space-based optical detectors, or ground-based infrared detectors equipped with laser guide star adaptive optics systems, a handful of progenitor stars have now either been directly identified or (more commonly) had upper luminosity limits established through non-detections. These studies enable direct comparisons with stellar evolution models that, in turn, permit estimates of the progenitor stars' physical characteristics to be made. I will review current progenitor characteristics (or constraints) inferred from this work for each of the major core-collapse supernova types (II-Plateau, II-Linear, IIb, IIn, Ib/c), with a particular focus on SN 2005gl, a type IIn event that is shown to have had an extremely luminous -- and thus very massive -- progenitor that likely exploded shortly after a violent, luminous blue variable-like eruption phase, contrary to standard theoretical predictions.

Red supergiants (RSGs) are a He-burning phase in the evolution of moderately massive stars (10-25 solar masses). For many years, the assumed physical properties of these stars placed them at odds with the predictions of evolutionary theory. We have recently determined new effective temperatures and luminosities for the RSG populations of galaxies with a wide range of metallicities, including the Milky Way, the Magellanic Clouds, and M31. We find that these new parameters greatly improve the agreement between the RSGs and the evolutionary tracks, although there are still notable difficulties with modeling the physical properties of the RSG population at lower metallicities. We have also examined the dust production rates and circumstellar dust properties of these stars, and consider how these might vary with metallicity. In addition, we have discovered a sample of four RSGs in the Magellanic Clouds that show considerable variations in their physical parameters, most notably their effective temperatures. In all four cases, when these stars are at their warmest they are also brighter, dustier, and more luminous. This behavior can be connected with sporadic dust production from these stars in their coolest states. We believe that these stars are undergoing an unstable evolutionary phase not previously associated with RSGs.

I will present modeling research work of the winds and circumstellar environments of a variety of prototypical hot and cool massive stars using advanced radiative transfer calculations. This research aims at unraveling the detailed physics of various mass-loss mechanisms of luminous stars in the upper portion of the HR-diagram.

Very recent 3-D radiative transfer calculations, combined with hydrodynamic simulations, show that the radiatively driven winds of OB Supergiants are structured due to large-scale density- and velocity-fields caused by rotating bright spots at the stellar equator. The mass-loss rates computed from matching Discrete Absorption Components in IUE observations of HD 64760 (B Ib) do not reveal appreciable changes from the rates of unstructured (smooth) wind models.

Intermediate yellow supergiants (such as the Yellow Hypergiant Rho Cas, F Ia0), on the other hand, show prominent spectroscopic signatures of strongly increased mass-loss rates during episodic outbursts caused by dramatic changes of the stellar photospheric conditions. Long-term high-resolution spectroscopic monitoring of cool hypergiants near the Yellow Evolutionary Void reveals that their mass-loss rates and wind-structure are dominated by photospheric eruptions and pulsations that impart mechanical momentum to the circumstellar environment by propagating acoustic (shock) waves.

In massive Red Supergiants, however, clear evidence for mechanical wave propagation from the sub-photospheric convection zones is not evident, despite their frequently observed spectroscopic and photometric variability. Recent spatially resolved HST-STIS observations inside Betelgeuse's very extended chromosphere and dust envelope show evidence of warm chromospheric gas far beyond the dust condensation radius of radiative transfer models. Models for these long-term spectroscopic observations demonstrate that the chromospheric pulsations are not spherically symmetric, similar to low-order non-radial oscillations in hot stars. These STIS observations point to the importance of mechanical wave propagation to heat and sustain chromospheric conditions in the extended dust-driven winds of red supergiants.

The galaxies of the Local Group provide an astrophysical laboratory where we can test theories of stellar evolution as a function of metallicity. But, for these tests to be meaningful, we must have a good understanding of the completeness of various samples. This talk will discuss what we do (and don't) know about the numbers of OB stars, Wolf-Rayets, Luminous Blue Variables, F and G Supergiants, and Red Supergiants in the Milky Way and other galaxies of the Local Group, and draw some illuminating comparisons between their numbers and what current massive star evolutionary theory predicts.

We have identified 17 new emission-line stars in the Galactic center region using NIR spectroscopy. The sources were selected for their hard X-ray emission, detected with the Chandra X-ray Observatory. The confirmed stars span a wide range of massive-star evolutionary stages, including OIa, Ofpe, WNh, WN, WNE, and WCd. We suspect that the hard X-ray emission is generated from the collision of supersonic stellar winds in massive binaries. The majority of the confirmed massive stars have no obvious association with known stellar clusters, but some may have been ejected or tidally stripped from clusters. The rarity of hard X-ray emitters among massive star populations suggests the existence of a large undetected population of massive stars in the Galactic center region.

Supernova 2008D was discovered serendipitously in January 2008 with the NASA Swift satellite via its X-ray emission and has generated great interest by astronomers (10 papers and counting). It is a supernova of Type Ib, that is, a core-collapse supernova whose massive stellar progenitor had been been stripped of most if not all of its outermost hydrogen layer, but had retained its next-inner helium layer, before explosion. We will present extensive optical and near-infrared light-curves and spectra (ranging from 17 hours to 109 days after the X-ray outburst), as well as our analysis of the Swift UV/optical data and of of the initial Swift X-ray outburst which we interpret as due to shock breakout emission. We will discuss the significance of this supernova, the derived size of its progenitor, the geometry of the explosion, and its implications for the supernova-GRB connection.

The new generation of optical and infrared interferometers have unprecedented angular resolution and sensitivity. I will review recent advances in studies of the upper HR diagram, including fundamental parameters of single stars, first interferometric imaging of interacting binaries, and new views of mass loss.

Normal unevolved early OB-type stars of masses ~8-20 Msun are the objects with the simplest photospheric physics in the upper part of the Hertzsprung-Russell diagram. They are described well by classical model atmospheres, unaffected by e.g. stellar winds like the hotter stars or by convection or chromospheres like the cool stars. However, their spectral analysis turned out to provide inconclusive results in the past decades, i.e. large uncertainties in basic stellar parameters and an overall enormous range in derived elemental abundances,challenging predictions from stellar and Galactic evolution models.

In order to improve the quantitative spectral analysis of such kind of star we have exhaustively updated their spectral modeling by constructructing robust model atoms for NLTE line-formation calculations. In parallel, we have implemented a self-consistent iterative spectral analysis technique, which brings numerous spectroscopic parameter and abundanceindicators in the optical and near-IR into agreement simultaneously.

Our efforts provided highly-promising results so far, i.e. a drastical reduction of statistical and systematic uncertainties in stellar parameters and chemical abundances. As a first application, a sample of stars from OB associations and the field in the solar neighbourhood covering a broad parameter range was analysed. The sample turned out to be chemically homogeneous on the ~10% level, in excellent accordance with results from published analyses of the ISM gas-phase. The abundances are consistent with published data on the Orion nebula. Our results have an immediate impact on several fields of contemporary astrophysics, i.e. stellar and Galactic chemical evolution models, they constrain the dust-phase composition of the local ISM, provide an independent view on the discussion of photospheric solar abundances and helioseismic constraints, and define the initial chemical composition for models of star and planet format ion in the solar neighbourhood.

Mass loss from hot, massive stars can occur both through steady winds of OB and WR phases, and through relatively brief eruptions during their Luminous Blue Variable (LBV) phase. This talk reviews how radiative momentum associated with the extreme luminosity of such stars is tapped to drive their mass loss. For OB stars, the steady outflows seem well described by the classical theory of Castor, Abbott, and Klein (CAK) for scattering of continuum radiation from a stellar core by line-transitions of metal ions in an otherwise optically thin wind. For WR stars, the winds themselves become optically thick, leading to ionization shifts and mass loss that can exceed the single scattering limit, with driving now arising from a complex combination of continuum and line opacity. In LBVs, the mass loss can be even more extreme, with the mechanical energy even becoming comparable to the radiative luminosity, for example in giant eruptions when the star's can actually exceed the classical Eddington limit. A key theme of this review is to compare and contrast the nature of radiative driving in these various stages of massive star evolution.

I will review current classical models of cool supergiant atmospheres, with emphasis on input data, and on the limits of used approximations. I will show comparisons to observed spectra. I will sketch the new prospects opened by the development of radiative-hydrodynamical models.

One of the most powerful tests of stellar evolution theory for massive stars is to observationally establish the causal mapping between different populations of massive stars and their explosions. This connection has been firmly proven only for a handful of objects, most notably in the case of supernova 1987A in the LMC with its blue-supergiant progenitor star Sk -69 202. However, the progenitors of most supernova types have not been identified directly. I will present two supernova-like transients discovered this year in the nearby galaxies NGC 6946 and NGC 300 for which we have identified the progenitors as dust-enshrouded massive stars in Spitzer images. This new class of luminous transients has progenitors that are extremely rare compared to known massive stellar populations identified in mid-IR observations of M33. I will discuss the implications of these findings in the context of "low-mass" massive stars (super-AGB stars) and connect it to electron-capture supernovae.

Massive stars spend a short time in between the hot and cool extremes of the HRD as tepid supergiants (Teff~8000-15000K). Possible scenarios are the first crossing from blue to red after the main sequence phase, blue loop episodes or a final crossing from red to blue in a late evolution stage. In consequence, important observational constraints on the evolution of massive stars can be derived from studies of tepid supergiants, providing valuable clues for future refinements of evolutionary models.

We discuss results from the analysis of a larger sample of Galactic objects. Recent improvements in the NLTE modelling of the supergiants' atmospheres and in the analysis methodology allow stellar parameters and elemental abundances to be determined with unprecedented accuracy from high-S/N, high-resolution spectra. In particular, for chemical abundances a reduction of uncertainties to as low as ~10-20% is achieved, facilitating the derivation of highly meaningful constraints. Observed abundances of the light elements (He, CNO) trace the transport efficiency of nuclear-processed material to the stellar surface, either by rotational mixing or during the first dredge-up. A mixing efficiency higher by about a factor 2 than predicted by current evolution models for rotating stars is indicated, implying that additional effects need to be considered in evolutionary models like e.g. the interplay of circulation and magnetic fields. Moreover, the observed abundance patterns allow stars after the first dredge-up to be identified, facilitating the extent of blue loops in the HRD to be mapped. Blue loops are found to extend to higher masses and to higher Teff than predicted by the current generation of stellar evolution models.

I will discuss the role that mass loss plays in the pre-SN evolution of massive stars in a variety of different scenarios, and what observable effect it may have on the resulting SN. The amount of mass lost, its speed, and how soon before core collapse the material is lost can have a dramatic effect on the resulting SN lightcurve and spectrum. Massive stars also trek across the HR diagram as they evolve, and the SN can look very different depending on where along this path core collapse occurs; it may not depend solely on initial mass. The most extreme pre-SN mass ejections in massive luminous blue variables (LBVs) have recently (and surprisingly) been linked to the very luminous Type IIn supernovae with extremely strong circumstellar interaction that dominates the spectrum and enhances the visual luminosity. In some cases these objects require strong LBV-like shell ejections in the decades immediatley before a SN. Strong winds or episodic mass loss of luminous red supergiants and yellow hypergiants may also lead to less extreme Type IIn events. Post-RSG blue supergiants like SN1987A's progenitor and lower-luminosity LBVs like HD168625 are also candidates for type II/IIb SNe with visible circumstellar material. Finally, progenitors that successfully shed their H envelopes (either through LBV eruptions, strong winds, or binary mass transfer) appear as Type Ib or Ic supernovae, some of which also show evidence for immediate pre-SN shell ejections. Many of the potential progenitors of Types Ib, Ic, IIn, IIb, and II-L overlap in their range of probable initial mass, and I will point to some open questions about how they fit together, and the roles of mass loss and initial mass in determining their relative rates.

M33 X-7 is the first stellar-mass black hole (BH) to be discovered in aneclipsing binary system. The high masses of the two components (~15 and ~70 solar masses for the BH and companion star, respectively), the short orbital period ( ~3.5 days), and the surprisingly low luminosity of the stellar companion make this system a challenge for the standard BH X-ray binary formation channels. We explore solutions to the current evolutionary state of this system. X-ray emission is due to wind accretion, since in the case of Roche-lobe overflow, the extreme mass ratio would lead to dynamically unstable mass transfer. We run single star evolutionary models of different masses, accounting for wind mass loss, to identify models that, at some evolutionary stage, satisfy the observational constraints on the effective temperature and luminosity of the companion star. Given the mass of the successful evolutionary models, we recalculate the BH mass from the observed mass function and inclination, the orbital properties of the binary, and its mean X-Ray luminosity. We use four different stellar evolution codes, modified to include a variety of current stellar wind prescriptions, and we account for systematic errors in radius, luminosity and effective temperature inherent to stellar evolution codes. Our best model, satisfying simultaneously the observational constraints on the donor star’s luminosity and effective temperature, and the mean X-ray luminosity, comprises of a ~13.4 solar masses BH and a ~54.0 solar masses companion star. We further examine the past orbital evolution of the binary and understand how the competing effects of wind mass loss and tidal interactions can lead to the formation of such an intriguing system.

The correlated stellar and nebular properties of coeval massive young clusters as a function of age will be emphasized. The recently investigated object Westerlund 1 is an essential addition to the standard sequence, which is unique in the solar vicinity. Its main-sequence turnoff is likely just below the Humphreys-Davidson Limit, resulting in a fully populated supergiant sequence across the entire Hertzsprung-Russell Diagram. In contrast, the comparably rich Scorpius OB1 association has no supergiants later than type B1.5; its turnoff mass must be just above the HD Limit. Between them, these two clusters define the location of the Limit to within about 10 Mo at the local metallicity.

In two-stage starbursts, two sequential cluster age phases with a difference of 1-2 Myr coexist. On the basis of detailed age calibrations in the Large Magellanic Cloud, the morphology of giant H II regions (or the lack thereof) accurately estimates the ages of extended regions in starburst galaxies. For instance, a monotonic age progression among five distinct young regions can be recognized across the face of the Antennae in HST images.

In this contributed talk I will present the observational evidence for supernova (SN) explosions of stars in the luminous blue variable (LBV) and the immediate post-LBV evolutionary phases. We now have compelling indications that two recent SNe of Type II-"narrow" (IIn) were the explosions of LBVs, including the direct identification of the progenitor LBV for one of these examples. A recent SN of Type Ic exploded as a helium star, two years after the powerful LBV outburst of its progenitor. These cases will likely have been discussed by other presenters at this Workshop in some detail. I will instead focus more on another example, SN 2001em, which was first identified as a Type Ib/c, but later evolved to Type IIn. I will argue that the progenitor of this SN exploded as a Wolf-Rayet (WR) star, following an eruptive LBV phase. Furthermore, I will show that two "SN impostors," i.e., extragalactic massive stars observed to undergo pre-SN LBV eruptions (similar to eta Carinae), may well have evolved to the WR phase in real time.

Massive stars becoming red supergiants lose a significant amount of their mass during that brief evolutionary phase. They then either explode as a hydrogen-rich supernova (SN Type II), or continue to evolve as a hotter supergiant (before exploding). The slow, dusty ejecta of the red supergiant will be over-run by the hot star wind and/or SN ejecta. I will present estimates of the conditions for this interaction and discuss some of the implications.

Massive stars throughout the HR diagram are often surrounded by large amounts of circumstellar material ejected through quiescent mass loss or violent events. The spatial distribution and composition of this material give clues about the nature of the mechanism(s) that are responsible for this circumstellar material, and the evolutionary phase of the underlying star. The talk will focus on similarities and differences in the circumstellar dust surrounding hot and cool massive stars.

We discuss the properties of Type Ib/c supernova progenitors in close binary systems using new evolutionary models of massive binary stars.We employed the most up-to-date estimate for the Wolf-Rayet mas loss rate, which is smaller by almost an order-of-magnitude than previously used in the similar studies by Woosley, Langer & Weaver (1995) and Wellstein & Langer (1999). We also considered the effect of rotation on the stellar structure, the transport of angular momentum due to rotationally induced hydrodynamic instabilities and the Spruit-Tayler dynamo, tidal interaction and exchange of mass and angular momentum. We find that, like massive single stars, most SN Ib/c progenitors in binary systems may retain enough angular momentum to produce millisecond pulsars, but not enough to produce magnetars or long gamma-ray bursts. SN Ib/c progenitors in binary systems are predicted to have a wide range of final masses even up to 10~Msun with helium envelopes of 0.2 - 1.5 Msun. These models indicate that, if lack of helium lines were due to small amounts of helium, most Type Ic supernovae should be produced from rather massive progenitors that may leave black holes as remnants. This may not be easily accommodated to current theories on supernova explosions and raises a question on the exact nature of Type Ic supernovae. A thin hydrogen layer 0.001-0.01 Msun is expected to be present in many SN Ib/c progenitors at the presupernova stage, in agreement with recent observational studies. The presence of hydrogen, together with a rather thick helium envelope, can lead to enormous expansion of a SN Ib/c progenitor with a final mass smaller than 5~Msun, even up to R = ~ 80 Rsun by the time of supernova explosion. This may have important consequences in shock break-outs and supernova light curves.

M33 X-7 is the first stellar-mass black hole (BH) to be discovered in an eclipsing binary system. The high masses of the two components (~15.65 and ~70.0 solar masses for the BH and companion star respectively) and the short orbital period ( ~3.45 days) make this system a challenge for the standard BH X-ray binary (XRB) formation channels. Given that the observed stellar luminosity is lower than what is expected from evolutionary models for a 70.0 solar mass star, we explore whether the system's properties can be explained with a less massive companion star. We assume that the X-ray emission is driven by the stellar wind of the companion, since in the case of Roche-lobe overflow, the extreme mass ratio of the binary would lead to dynamically unstable mass transfer. We use four different stellar evolution codes, modified to include a full spectrum of stellar wind prescriptions, to account for systematic errors in radius, luminosity and effective temperature inherent to stellar evolution codes. Our best model, satisfying simultaneously the observational constraints on the donor star’s luminosity and effective temperature, and the mean X-ray luminosity, comprises of a ~13.4 solar mass BH and a ~54.0 solar mass companion star. The next step in our analysis is to unveil the past orbital evolution of the binary and understand how the competing effects of wind mass loss and tidal interactions can lead to the formation of such an extraordinary system.


Infrared Processing and Analysis CenterJet Propulsion LaboratoryNASA