arXiv subjects

G. M. Harper

Publications and source records attributed to G. M. Harper.

At least 19 recordsLinked to original sources

arXiv2026

ALMA high resolution observations of Betelgeuse: Persistent structure spanning the inner atmosphere

The extended atmosphere of red supergiants (RSGs) forms an important link in the process of mass loss and the subsequent enrichment of the interstellar medium. Large-scale convection is thought to play a significant role, which is likely to result in irregularities in the surface. High resolution, high contrast sub-mm images of Betelgeuse - one of the closest RSGs - are used to probe the structure and temporal stability of the inner 1-2$R_\star$ of its atmosphere. Using ALMA in the longest baseline configuration, continuum emission and lines of SiO and CO and their isotopomers were observed at $\lambda$0.6-1.4mm, giving beamwidths down to 7mas at the shortest wavelengths. These were compared with a similar observation taken at 0.9mm approximately 7 years earlier. The observed continuum emission arises mostly from an optically-thick mm/sub-mm photosphere of radius 1.1-1.3$R_\star$ with a relatively constant temperature of $\sim$2300K, but with two hotter patches to the NE and SW. The brightest of these has a temperature enhancement of $\sim$800K, and its location and intensity appears relatively unchanged since the 2015 observation. The sub-mm photosphere shows deviations of up to $\pm$ 6% in radius, with weaker continuum extending out to $\sim$2.5$R_\star$ - similar to the extent of clumpy emission in SiO and CO. The hot regions of gas and deviations from radial symmetry are thought to be associated with active shocks driven by underlying convective cells, although their lifetimes appear longer than model predictions. They lie near the proposed poles of the star, which might suggest enhanced and relatively stable polar convection. The present data show no clear evidence for stellar rotation in the extended line emission or absorption against the photosphere, although the structure of the gas emission has changed significantly since 2015.

astro-ph.SR↗

arXiv2020

ALMA and VLA reveal the lukewarm chromospheres of the nearby red supergiants Antares and Betelgeuse

We first present spatially resolved ALMA and VLA continuum observations of the early-M red supergiant Antares to search for the presence of a chromosphere at radio wavelengths. We resolve the free-free emission of the Antares atmosphere at 11 unique wavelengths between 0.7 mm (ALMA band 8) and 10 cm (VLA S band). The projected angular diameter is found to continually increase with increasing wavelength, from a low of 50.7 mas at 0.7 mm up to a diameter of 431 mas at 10 cm, which corresponds to 1.35 and 11.6 times the photospheric angular diameter, respectively. All four ALMA measurements show that the shape of the atmosphere is elongated, with a flattening of 15% at a similar position angle. The disk-averaged gas temperature of the atmosphere initially rises from a value of 2700 K at 1.35 $R_{\star}$ (i.e., 0.35 $R_{\star}$ above the photosphere) to a peak value of 3800 K at $\sim$2.5 $R_{\star}$, after which it then more gradually decreases to 1650 K at 11.6 $R_{\star}$. The rise in gas temperature between 1.35 $R_{\star}$ and $\sim$2.5 $R_{\star}$ is evidence for a chromospheric temperature rise above the photosphere of a red supergiant. We detect a clear change in the spectral index across the sampled wavelength range, with the flux density ${S_ν} \propto ν^{1.42}$ between 0.7 mm and 1.4 cm, which we associate with chromosphere-dominated emission, while the flux density ${S_ν} \propto ν^{0.8}$ between 4.3 cm and 10 cm, which we associate with wind-dominated emission. We then perform nonlocal thermal equilibrium modeling of the far-ultraviolet radiation field of another early-M red supergiant, Betelgeuse, and find that an additional hot (i.e., $>7000$ K) chromospheric photoionization component with a much smaller filling factor must also exist throughout the chromospheres of these stars.

astro-ph.SR↗

arXiv2019

NOEMA maps the CO $J = 2-1$ environment of the red supergiant $μ$ Cep

Red supergiant stars are surrounded by a gaseous and dusty circumstellar environment created by their mass loss which spreads heavy elements into the interstellar medium. The structure and the dynamics of this envelope are crucial to understand the processes driving the red supergiant mass loss and the shaping of the pre-supernova ejecta. We have observed the emission from the CO $J = 2-1$ line from the red supergiant star $μ$~Cep with the NOEMA interferometer. In the line the synthesized beam was $0.92 \times 0.72$~arcsec ($590 \times 462$~au at 641~pc). The continuum map shows only the unresolved contribution of the free-free emission of the star chromosphere. The continuum-subtracted channel maps reveal a very inhomogeneous and clumpy circumstellar environment. In particular, we detected a bright CO clump, as bright as the central source in the line, at 1.80~arcsec south-west from the star, in the blue channel maps. After a deprojection of the radial velocity assuming two different constant wind velocities, the observations were modelled using the 3D radiative transfer code \textsc{lime} to derive the characteristics of the different structures. We determine that the gaseous clumps observed around $μ$~Cep are responsible for a mass loss rate of $(4.9 \pm 1.0) \times 10^{-7}~{\rm M}_\odot\,{\rm yr}^{-1}$, in addition to a spatially unresolved wind component with an estimated mass-loss rate of $2.0 \times 10^{-6}~{\rm M}_\odot\,{\rm yr}^{-1}$. Therefore, the clumps have a significant role in $μ$~Cep's mass loss ($\ge 25 \%$). We cannot exclude that the unresolved central outflow may be made of smaller unresolved clumps.

astro-ph.SR↗

arXiv2017

The inhomogeneous sub-millimeter atmosphere of Betelgeuse

The mechanisms responsible for heating the extended atmospheres of early-M spectral-type supergiants are poorly understood. So too is the subsequent role these mechanisms play in driving the large mass-loss rates of these stars. Here we present ALMA long (i.e., $\sim$16 km) baseline 338 GHz (0.89 mm) continuum observations of the free-free emission in the extended atmosphere of the M2 spectral-type supergiant Betelgeuse. The spatial resolution of 14 mas exquisitely resolves the atmosphere, revealing it to have a mean temperature of 2760 K at $\sim$1.3 R$_{\star}$, which is below both the photospheric effective temperature ($T_{\textrm{eff}} = 3690$ K) and the temperatures at $\sim$$2 $R$_{\star}$. This is unambiguous proof for the existence of an inversion of the mean temperature in the atmosphere of a red supergiant. The emission is clearly not spherically symmetric with two notable deviations from a uniform disk detected in both the images and visibilities. The most prominent asymmetry is located in the north-east quadrant of the disk and is spatially resolved showing it to be highly elongated with an axis-ratio of 2.4 and occupying $\sim$$5\%$ of the disk projected area. Its temperature is approximately 1000 K above the measured mean temperature at 1.3 R$_{\star}$. The other main asymmetry is located on the disk limb almost due east of the disk center and occupies $\sim$$3\%$ of the disk projected area. Both emission asymmetries are clear evidence for localized heating taking place in the atmosphere of Betelgeuse. We suggest that the detected localized heating is related to magnetic activity generated by large-scale photospheric convection.

astro-ph.SR↗

arXiv2017

An Updated 2017 Astrometric Solution for Betelgeuse

We provide an update for the astrometric solution for the Type II supernova progenitor Betelgeuse using the revised Hipparcos Intermediate Astrometric Data (HIAD) of van Leeuwen, combined with existing VLA and new e-MERLIN and ALMA positions. The 2007 Hipparcos refined abscissa measurements required the addition of so-called Cosmic Noise of 2.4 mas to find an acceptable 5-parameter stochastic solution. We find that a measure of radio Cosmic Noise should also be included for the radio positions because surface inhomogeneities exist at a level significant enough to introduce additional intensity centroid uncertainty. Combining the 2007 HIAD with the proper motions based solely on the radio positions leads to a parallax of $π= 5.27\pm 0.78$ mas ($190^{+33}_{-25}$ pc), smaller than the Hipparcos 2007 value of $6.56\pm 0.83$ mas ($152^{+22}_{-17}$ pc; van Leeuwen 2007). Furthermore, combining the VLA and new e-MERLIN and ALMA radio positions with the 2007 HIAD, and including radio Cosmic Noise of 2.4 mas, leads to a nominal parallax solution of $4.51 \pm 0.80$ mas ($222^{+48}_{-34}$ pc), which while only $0.7σ$ different from the 2008 solution of Harper et al. it is $2.6σ$ different from the solution of van Leeuwen. An accurate and precise parallax for Betelgeuse is always going to be difficult to obtain because it is small compared to the stellar angular diameter ($θ=44$ mas). We outline an observing strategy, utilizing future mm and sub-mm high-spatial resolution interferometry that must be used if substantial improvements in precision and accuracy of the parallax and distance are to be achieved.

astro-ph.SR↗

arXiv2015

SOFIA/EXES Observations of Water Absorption in the Protostar AFGL 2591 at High Spectral Resolution

We present high spectral resolution (~3 km/s) observations of the nu_2 ro-vibrational band of H2O in the 6.086--6.135 micron range toward the massive protostar AFGL 2591 using the Echelon-Cross-Echelle Spectrograph (EXES) on the Stratospheric Observatory for Infrared Astronomy (SOFIA). Ten absorption features are detected in total, with seven caused by transitions in the nu_2 band of H2O, two by transitions in the first vibrationally excited nu_2 band of H2O, and one by a transition in the nu_2 band of H2{18}O. Among the detected transitions is the nu_2 1(1,1)--0(0,0) line which probes the lowest lying rotational level of para-H2O. The stronger transitions appear to be optically thick, but reach maximum absorption at a depth of about 25%, suggesting that the background source is only partially covered by the absorbing gas, or that the absorption arises within the 6 micron emitting photosphere. Assuming a covering fraction of 25%, the H2O column density and rotational temperature that best fit the observed absorption lines are N(H2O)=(1.3+-0.3)*10^{19} cm^{-2} and T=640+-80 K.

astro-ph.GA↗

arXiv2014

ALMA Observations of Anisotropic Dust Mass-loss in the Inner Circumstellar Environment of the Red Supergiant VY Canis Majoris

The processes leading to dust formation and the subsequent role it plays in driving mass loss in cool evolved stars is an area of intense study. Here we present high resolution ALMA Science Verification data of the continuum emission around the highly evolved oxygen-rich red supergiant VY CMa. These data enable us to study the dust in its inner circumstellar environment at a spatial resolution of 129 mas at 321 GHz and 59 mas at 658 GHz, thus allowing us to trace dust on spatial scales down to 11 R$_{\star}$ (71 AU). Two prominent dust components are detected and resolved. The brightest dust component, C, is located 334 mas (61 R$_{\star}$) South East of the star and has a dust mass of at least $2.5\times 10^{-4}$ M$_{\odot}$. It has a dust emissivity spectral index of $β=-0.1$ at its peak, implying that it is optically thick at these frequencies with a cool core of $T_{d}\lesssim 100$ K. Interestingly, not a single molecule in the ALMA data has emission close to the peak of this massive dust clump. The other main dust component, VY, is located at the position of the star and contains a total dust mass of $4.0 \times 10^{-5} $M$_{\odot}$. It also contains a weaker dust feature extending over $60$ R$_{\star}$ to the North with the total component having a typical dust emissivity spectral index of $β=0.7$. We find that at least $17\%$ of the dust mass around VY CMa is located in clumps ejected within a more quiescent roughly spherical stellar wind, with a quiescent dust mass loss rate of $5 \times 10^{-6}$ M$_{\odot} $yr$^{-1}$. The anisotropic morphology of the dust indicates a continuous, directed mass loss over a few decades, suggesting that this mass loss cannot be driven by large convection cells alone.

astro-ph.SR↗

arXiv2014

ALMA sub-mm maser and dust distribution of VY Canis Majoris

Cool, evolved stars have copious, enriched winds. The structure of these winds and the way they are accelerated is not well known. We need to improve our understanding by studying the dynamics from the pulsating stellar surface to about 10 stellar radii, where radiation pressure on dust is fully effective. Some red supergiants have highly asymmetric nebulae, implicating additional forces. We retrieved ALMA Science Verification data providing images of sub-mm line and continuum emission from VY CMa. This enables us to locate water masers with milli-arcsec precision and resolve the dusty continuum. The 658-, 321- and 325-GHz masers lie in irregular, thick shells at increasing distances from the centre of expansion. For the first time this is confirmed as the stellar position, coinciding with a compact peak offset to the NW of the brightest continuum emission. The maser shells (and dust formation zone) overlap but avoid each other on tens-au scales. Their distribution is broadly consistent with excitation models but the conditions and kinematics appear to be complicated by wind collisions, clumping and asymmetries.

astro-ph.SR↗

arXiv2014

Systematic trend of water vapour absorption in red giant atmospheres revealed by high resolution TEXES 12 micron spectra

The structures of the outer atmospheres of red giants are very complex. The notion of large optically thick molecular spheres around the stars (MOLspheres) has been invoked in order to explain e.g. spectro-interferometric observations. However, high-resolution spectra in the mid-IR do not easily fit into this picture. They rule out any large sphere of water vapour in LTE surrounding red giants. Our aim here is to investigate high-resolution, mid-infrared spectra for a range of red giants, from early-K to mid M. We have recorded 12 microns spectra of 10 well-studied bright red giants, with TEXES on the IRTF. We find that all giants in our study cooler than 4300 K, spanning a range of effective temperatures, show water absorption lines stronger than expected. The strengths of the lines vary smoothly with spectral type. We identify several spectral features in the wavelength region that undoubtedly are formed in the photosphere. From a study of water-line ratios of the stars, we find that the excitation temperatures, in the line-forming regions, are several hundred Kelvin lower than expected from a classical photospheric model. This could either be due to an actually lower temperature structure in the outer regions of the photospheres caused by, for example, extra cooling, or due to non-LTE level populations, affecting the source function and line opacities. We have demonstrated that these diagnostically interesting water lines are a general feature of red giants across spectral types, and we argue for a general explanation of their formation rather than explanations requiring specific properties. Since the water lines are neither weak (filled in by emission) nor appear in emission, as predicted by LTE MOLsphere models in their simplest forms, the evidence for the existence of such large optically-thick, molecular spheres enshrouding the stars is weakened. (abbreviated)

astro-ph.SR↗

arXiv2014

Evolved stars with complex atmospheres - the high spectral resolution, mid-IR view

The physical structures of the outer atmospheres of red giants are not known. They are certainly complex and a range of recent observations are showing that we need to embrace to non-classical atmosphere models to interpret these regions. This region's properties is of importance, not the least, for the understanding of the mass-loss mechanism for these stars, which is not still understood. Here, we present observational constraints of the outer regions of red giants, based on mid-IR, high spectral resolution spectra. We also discuss possible non-LTE effects and highlight a new non-LTE code that will be used to analyse the spectra of these atmospheric layers. We conclude by mentioning our new SOFIA/EXES observations of red giants at 6 microns, where the vibration-rotation lines of water vapour can be detected and spectrally resolved for the first time.

astro-ph.SR↗

arXiv2014

Fluorine in the solar neighborhood - is it all produced in AGB-stars?

The origin of 'cosmic' fluorine is uncertain, but there are three proposed production sites/mechanisms: AGB stars, $ν$ nucleosynthesis in Type II supernovae, and/or the winds of Wolf-Rayet stars. The relative importance of these production sites has not been established even for the solar neighborhood, leading to uncertainties in stellar evolution models of these stars as well as uncertainties in the chemical evolution models of stellar populations. We determine the fluorine and oxygen abundances in seven bright, nearby giants with well-determined stellar parameters. We use the 2.3 $μ$m vibrational-rotational HF line and explore a pure rotational HF line at 12.2 $μ$m. The latter has never been used before for an abundance analysis. To be able to do this we have calculated a line list for pure rotational HF lines. We find that the abundances derived from the two diagnostics agree. Our derived abundances are well reproduced by chemical evolution models only including fluorine production in AGB-stars and therefore we draw the conclusion that this might be the main production site of fluorine in the solar neighborhood. Furthermore, we highlight the advantages of using the 12 $μ$m HF lines to determine the possible contribution of the $ν$-process to the fluorine budget at low metallicities where the difference between models including and excluding this process is dramatic.

astro-ph.GA↗

arXiv2014

The Chemical Evolution of Fluorine in the Bulge - High-resolution K-band spectra of giants in three fields

Possible main formation sites of F in the Universe include AGB stars, the ν-process in Type II SNe, and/or W-R stars. The importance of the W-R stars has theoretically been questioned and they are probably not needed in the modelling of the chemical evolution of F in the solar neighborhood. It has, however, been suggested that W-R stars are indeed needed to explain the chemical evolution of F in the Bulge. The molecular spectral data of the often used HF-molecule has not been presented in a complete and consistent way and has recently been debated in the literature. In this article we determine the [F/O] vs. [O/H] trend in the Bulge to investigate the possible contribution from W-R stars. Additionally, we present here a HF line list for the K- and L-bands (including the often used 23358.33 Å line) and an accompanying partition function. The F abundances were determined using spectral fitting from hi-res NIR spectra of eight K giants recorded by the spectrograph CRIRES. We have also re-analyzed five previously published Bulge giants using our new HF molecular data. We find that the F-O abundance in the Bulge probably cannot be explained with chemical evolution models including only AGB-stars and the ν-process in SNe Type II, i.e. a significant amount of F production in W-R stars is likely needed to explain the F abundance in the Bulge. Concerning the HF line list, we find that a possible reason for the inconsistencies in the literature, with two different excitation energies being used, is two different definitions of the zero-point energy for the HF molecule and therefore also two accompanying different dissociation energies. Both line lists are correct, as long as the corresponding consistent partition function is used in the spectral synthesis. However, we suspect this has not been the case in several earlier works leading to F abundances 0.3 dex too high.

astro-ph.GA↗

arXiv2013

e-MERLIN resolves Betelgeuse at wavelength 5 cm

Convection, pulsation and magnetic fields have all been suggested as mechanisms for the transport of mass and energy from the optical photosphere of red supergiants, out to the region where the stellar wind is launched. We imaged the red supergiant Betelgeuse at 0.06-0.18 arcsec resolution, using e-MERLIN at 5.5--6.0 GHz, with a sensitivity of ~0.01 mJy/beam. Most of the radio emission comes from within an ellipse (0.235x0.218) arcsec^2 (~5x the optical radius), with a flux density of 1.62 mJy, giving an average brightness temperature ~1250 K. This radio photosphere contains two hotspots of 0.53 and 0.79 mJy/beam, separated by 90 milli-arcsec, with brightness temperatures 5400+/-600 K and 3800+/-500 K. Similar hotspots, at more than double the distance from the photosphere of those seen in any other regime, were detected by the less-sensitive `old' MERLIN in 1992, 1995 and 1996 and many exceed the photospheric temperature of 3600 K. Such brightness temperatures are high enough to emanate from pockets of chromospheric plasma. Other possibilities include local shock heating, the convective dredge-up of hot material or exceptionally cool, low density regions, transparent down to the hottest layer at ~40 milliarcsec radius. We also detect an arc 0.2--0.3 arcsec to the SW, brightness temperature ~150 K, in a similar direction to extensions seen on both smaller and larger scales in the infra-red and in CO at mm wavelengths. These preliminary results will be followed by further e-MERLIN, VLA and ALMA observations to help resolve the problem of mass elevation from 1 to 10 R* in red supergiants.

astro-ph.SR↗

arXiv2012

Chromospheric Thermal Continuum Millimetre Emission from non-dusty K and M Red Giants

We examine the thermal free-free millimetre fluxes expected from non-dusty and non-pulsating K through mid-M giant stars based on our limited understanding of their inhomogeneous chromospheres. We present a semi-analytic model that provides estimates of the radio fluxes for the mm wavelengths (e.g., CARMA, ALMA, JVLA Q-band) based on knowledge of the effective temperatures, angular diameters and chromospheric Mg II h & k emission fluxes. At 250 GHz, the chromospheric optical depths are expected to be significantly less than unity, which means that fluxes across the mm and sub-mm range will have a contribution from the chromospheric mate- rial that gives rise to the ultraviolet emission spectrum, as well as the cool molecular material known to exist above the photosphere. We predict a lower bound to the inferred brightness temperature of red giants based on heating at the basal-flux limit if the upper chromospheres have filling factor 1. Multi-frequency mm observations should provide important new information on the structuring of the inhomogeneous chromospheres, including the boundary layer, and allow tests of competing theoretical models for atmospheric heating. We comment on the suitability of these stars as mm flux calibrators.

astro-ph.SR↗

arXiv2009

TEXES Observations of M Supergiants: Dynamics and Thermodynamics of Wind Acceleration

We have detected [Fe II] 17.94 um and 24.52 um emission from a sample of M supergiants using TEXES on the IRTF. These low opacity emission lines are resolved at R = 50, 000 and provide new diagnostics of the dynamics and thermodynamics of the stellar wind acceleration zone. The [Fe II] lines, from the first excited term, are sensitive to the warm plasma where energy is deposited into the extended atmosphere to form the chromosphere and wind outflow. These diagnostics complement previous KAO and ISO observations which were sensitive to the cooler and more extended circumstellar envelopes. The turbulent velocities, Vturb is about 12 to 13 km/s, observed in the [Fe II] forbidden lines are found to be a common property of our sample, and are less than that derived from the hotter chromospheric C II] 2325 Angstrom lines observed in alpha Ori, where Vturb is about 17 to 19 km/s. For the first time, we have dynamically resolved the motions of the dominant cool atmospheric component discovered in alpha Ori from multi-wavelength radio interferometry by Lim et al. (1998). Surprisingly, the emission centroids are quite Gaussian and at rest with respect to the M supergiants. These constraints combined with model calculations of the infrared emission line fluxes for alpha Ori imply that the warm material has a low outflow velocity and is located close to the star. We have also detected narrow [Fe I] 24.04 um emission that confirms that Fe II is the dominant ionization state in alpha Ori's extended atmosphere.

astro-ph.SR↗

arXiv2008

Mg I emission lines at 12 and 18 micrometer in K giants

The solar Mg I emission lines at 12 micrometer have already been observed and analyzed well. Previous modeling attempts for other stars have, however, been made only for Procyon and two cool evolved stars, with unsatisfactory results for the latter. We present high-resolution observational spectra for the K giants Pollux, Arcturus, and Aldebaran, which show strong Mg I emission lines at 12 micrometer as compared to the Sun. We also present the first observed stellar emission lines from Mg I at 18 micrometer and from Al I, Si I, and presumably Ca I at 12 micrometer. To produce synthetic line spectra, we employ standard non-LTE modeling for trace elements in cool stellar photospheres. We compute model atmospheres with the MARCS code, apply a comprehensive magnesium model atom, and use the radiative transfer code MULTI to solve for the magnesium occupation numbers in statistical equilibrium. We successfully reproduce the observed Mg I emission lines simultaneously in the giants and in the Sun, but show how the computed line profiles depend critically on atomic input data and how the inclusion of energy levels with n > 9 and collisions with neutral hydrogen are necessary to obtain reasonable fits.

astro-ph↗

arXiv2006

From Radio to X-ray: The Quiescent Atmosphere of the dMe Flare Star EV Lacertae

We report on multi-wavelength observations spanning radio to X-ray wavelengths of the M dwarf flare star, EV Lacertae, probing the characteristics of the outer atmospheric plasma from the upper chromosphere to the corona. We detect the star at a wavelength of 2 cm (15 GHz) for the first time. UV and FUV line profiles show evidence of nonthermal broadening, and the velocity width appear to peak at lower temperatures than in the Sun; this trend is confirmed in another active M dwarf flare star. Electron density measurements indicate nearly constant electron pressures between $\log T=$5.2 and 6.4. At higher coronal temperatures, there is a sharp increase of two orders of magnitude in density (n$_{e}\sim10^{13}$ cm$^{-3}$ at $\log T=$6.9). X-ray, EUV, FUV and NUV spectra constrain the DEM from the upper chromosphere through the corona. The coronal pressures are inconsistent with the assumption of hydrostatic equilibrium, either through EM modeling or application of scaling laws, and imply large conductive loss rates and a large energy input at the highest temperatures. The timescales for radiative and conductive losses in EV Lac's upper atmosphere imply that significant continued heating must occur for the corona to maintain its quiescent properties. The high frequency radio detection requires the high temperature X-ray-emitting coronal plasma to be spatially distinct from the radio emission source. Length scales in the low-temperature corona are markedly larger than those in the high-temperature corona, further suggestions of an inhomogeneous mixture of thermal and nonthermal coronal plasma.

astro-ph↗

arXiv2006

Water vapor on supergiants. The 12 micron TEXES spectra of mu Cephei

Several recent papers have argued for warm, semi-detached, molecular layers surrounding red giant and supergiant stars, a concept known as a MOLsphere. Spectroscopic and interferometric analyses have often corroborated this general picture. Here, we present high-resolution spectroscopic data of pure rotational lines of water vapor at 12 microns for the supergiant mu Cephei. This star has often been used to test the concept of molecular layers around supergiants. Given the prediction of an isothermal, optically thick water-vapor layer in Local Thermodynamic Equilibrium around the star (MOLsphere), we expected the 12 micron lines to be in emission or at least in absorption but filled in by emission from the molecular layer around the star. Our data, however, show the contrary; we find definite absorption. Thus, our data do not easily fit into the suggested isothermal MOLsphere scenario. The 12 micron lines, therefore, put new, strong constraints on the MOLsphere concept and on the nature of water seen in signatures across the spectra of early M supergiants. We also find that the absorption is even stronger than that calculated from a standard, spherically symmetric model photosphere without any surrounding layers. A cool model photosphere, representing cool outer layers is, however, able to reproduce the lines, but this model does not account for water vapor emission at 6 microns. Thus, a unified model for water vapor on mu Cephei appears to be lacking. It does seem necessary to model the underlying photospheres of these supergiants in their whole complexity. The strong water vapor lines clearly reveal inadequacies of classical model atmospheres.

astro-ph↗