We describe observational evidence for a new kind of interacting-binary-star outburst that involves both an accretion instability and an increase in thermonuclear shell burning on the surface of an accreting white dwarf. We refer to this new type of eruption as a combination nova. In late 2000, the prototypical symbiotic star Z Andromedae brightened by roughly two magnitudes in the optical. We observed the outburst in the radio with the VLA and MERLIN, in the optical both photometrically and spectroscopically, in the far ultraviolet with F U SE, and in the X-rays with both Chandra and XM M . The two-year-long event had three distinct stages. During the first stage, the optical rise closely resembled an earlier, small outburst that was caused by an accretion-disk instability. In the second stage, the hot component ejected an optically thick shell of material. In the third stage, the shell cleared to reveal a white dwarf whose luminosity remained on the order of 10 4 L ⊙ for approximately one year. The eruption was thus too energetic to have been powered by accretion alone. We propose that the initial burst of accretion was large enough to trigger enhanced nuclear burning on the surface of the white dwarf and the ejection of an optically thick shell of material. This outburst therefore combined elements of both a dwarf nova and a classical nova. Our results have implications for the long-standing problem of producing shell flashes with short recurrence times on low-mass white dwarfs in symbiotic stars.

We have fit synthetic visibilities from three-dimensional (CO 5 BOLD+PHOENIX) and one-dimensional (PHOENIX, ATLAS 12) model stellar atmospheres of Procyon ( F5 IV) to high-precision interferometric data from the VLT Interferometer (K band) and from the Mark III interferometer (500 and 800 nm). These data sets provide a test of theoretical wavelength-dependent limb-darkening predictions. The work of Allende Prieto et al. has shown that the temperature structure from a spatially and temporally averaged three-dimensional hydrodynamic model produces significantly less limb darkening at 500 nm relative to the temperature structure of a one-dimensional MARCS model atmosphere with a standard mixing-length approximation for convection. Our direct fits to the interferometric data confirm this prediction. A one-dimensional ATLAS 12 model with ''approximate overshooting'' provides the required temperature gradient. We show, however, that one-dimensional models cannot reproduce the ultraviolet spectrophotometry below 160 nm with effective temperatures in the range constrained by the measured bolometric flux and angular diameter. We find that a good match to the full spectral energy distribution can be obtained with a composite model consisting of a weighted average of 12 one-dimensional model atmospheres based on the surface intensity distribution of a three-dimensional granulation simulation. We emphasize that one-dimensional models with overshooting may realistically represent the mean temperature structure of F-type stars such as Procyon, but the same models will predict redder colors than observed because they lack the multicomponent temperature distribution expected for the surfaces of these stars.

Aims. We present coordinated near-infrared K-band interferometric and optical spectroscopic observations of the M 1.5 giant α Cet (Menkar) obtained with the instruments VINCI and UVES at the Paranal Observatory. Spherically symmetric PHOENIX stellar model atmospheres are constrained by comparison to our interferometric and spectroscopic data, and high-precision fundamental parameters of Menkar are obtained. Methods. Our high-precision VLTI/VINCI observations in the first and second lobes of the visibility function directly probe the model-predicted strength of the limb darkening effect in the K-band and the stellar angular diameter. The high spectral resolution of UVES of R = 80 000-110 000 allows us to confront in detail observed and model-predicted profiles of atomic lines and molecular bands. Results. We show that our derived PHOENIX model atmosphere for Menkar is consistent with both the measured strength of the limbdarkening in the near-infrared K-band and the profiles of spectral bands around selected atomic lines and TiO bandheads from 370 nm to 1000 nm. At the detailed level of our high spectral resolution, however, noticeable discrepancies between observed and synthetic spectra exist. We obtain a high-precision Rosseland angular diameter of Θ Ross = 12.20 mas ± 0.04 mas. Together with the Hipparcos parallax of 14.82 mas ± 0.83 mas, it corresponds to a Rosseland radius of R Ross = 89 ± 5R , and together with the bolometric flux based on available spectrophotometry, to an effective temperature of T eff = 3795 K ± 70 K. The luminosity based on these values is L = 1460 L ± 300 L . Relying on stellar evolutionary tracks, these values correspond to a mass M = 2.3 M ± 0.2 M and a surface gravity log g = 0.9 ± 0.1 (cgs). Conclusions. Our approach illustrates the power of combining interferometry and high-resolution spectroscopy to constrain and calibrate stellar model atmospheres. The simultaneous agreement of the model atmosphere with our interferometric and spectroscopic data increases confidence in the reliability of the modelling of this star, while discrepancies at the detailed level of the high resolution spectra can be used to further improve the underlying model.

No abstract

The variation of the specific intensity across the stellar disc is an essential input parameter in surface brightness reconstruction techniques such as Doppler imaging, where the relative intensity contributions of different surface elements are important in detecting star‐spots. We use phoenix and atlas model atmospheres to model light curves derived from high precision (signal‐to‐noise ratio ≃ 5000) Hubble Space Telescope (HST) data of the eclipsing binary SV Cam (F9V+K4V), where the variation of specific intensity across the stellar disc will determine the contact points of the binary system light curve. For the first time, we use χ2 comparison fits to the first derivative profiles to determine the best‐fitting model atmosphere. We show the wavelength dependence of the limb darkening and that the first derivative profile is sensitive to the limb‐darkening profile very close to the limb of the primary star. It is concluded that there is only a marginal difference (<1σ) between the χ2 comparison fits of the two model atmospheres to the HST light curve at all wavelengths. The usefulness of the second derivative of the light curve for measuring the sharpness of the primary's limb is investigated, but we find that the data are too noisy to permit a quantitative analysis.