Systematic photometric monitoring of active RS CVn binaries carried out at Catania with the automatic telescope APT-80, is being complemented with Hα low- and high-resolution spectroscopy. The relationship between photospheric and chromospheric activity is investigated in order to shed light into the complex three-dimensional structure of surface activity. Preliminary results on the photometric and Hα monitoring of the active binaries, UX Ari, RS CVn, BM CVn, HK Lac, IM Peg, V 711 Tau, and EI Eri, are here reported.

Through examination of line profile asymmetries and their phase-dependent behavior at high-resolution and high S/N we can determine the spatial structure (i.e., size, location, temperature, magnetic field distribution) of atmospheric inhomogeneities on stars other than the Sun (Vogt & Penrod 1983; Walter et al. 1987; Neff et al. 1989). In an effort to simultaneously probe several lines in addition to Mg II we observed the RS CVn V711 Tau (HR 1099, Prot = 2.d8, G5 IV + K1 IV) with the GHRS in the fall of 1993. Ultraviolet observations were obtained with HST's GHRS over a one week interval beginning 1993 September 14. Full details can be found in Dempsey et al. (1996; see also Wood et al. 1995). In this paper we present results of detailed modeling of the the Mg II emission profiles. Line fluxes and profile models for the emission lines were calculated using least-squares gaussian fits as described in Neff et al. (1989).

Stars close to the main sequence, if they rotate rapidly enough, show magnetic activity in the form of spots, active regions, and strong chromospheres. With the exception of the hybrid stars, stars cooler and more luminous than the so-called Linsky-Haisch dividing line, in contrast, are not known to have any of these phenomena. In fact, it is not clear such stars, even though rotating and convective, show any magnetic phenomena at all. We discuss (1) photometry used to search for spots in the hybrid stars, which we have not found, and (2) evidence for chromospheric variation in K supergiants that takes the form of enhanced winds, but which may be analogous to the activity of stars closer to the main sequence.

We review the evidence, from rotational modulation, for association of plages and spots on active binaries. We find reasonably convincing support for such an association from IUE SWP observations of RS CVn stars, but little or no evidence from MgII line fluxes. There is a similar lack of evidence for BY Dra systems. The results can be interpreted as evidence for: (a) the presence of active longitudes which persist on some RS CVn stars for several years, (b) high filling factors in MgII emission and (c) lower filling factors, and therefore greater contrast, in SWP emission. There is a suggestion that the association between plages and spots is more clearly seen in longer period RS CVn stars than the very active, short period, systems favoured by IUE observers.

We discuss the theory of quasi-static coronal loops, introducing a phase plane representation to study loop solutions independently of specific boundary conditions. Emphasis is put on the effects of loop expansion, heat input and gravitational stratification on the differential emission measure, and on the intrinsic limitations of spectroscopic observations for deriving loop parameters. We show that certain classes of published loop solutions cannot actually exist. For expanding loops new classes of loop solutions, with rather special properties, are presented. Special attention is paid to loops in binary systems and on rapidly rotating stars.

Stellar coronae, like the interstellar medium, are multi-phase media. Physical conditions in different parts of the corona allow material to exist in thermal equilibrium either as a tenuous X-ray plasma at temperatures of several million degrees, or in a denser form at temperatures low enough for hydrogen to recombine.

I examine the possibility that large, cool prominences can be formed in close binary systems, when coronal structures grow beyond the point where centrifugal force balances gravitational attraction. X-ray light curves of the eclipsing binary XY UMa indicate that an extended corona is probably present and this is likely to reach heights > 1.5 R⊙, where centrifugal compression may render plasma in the coronal loop apexes unstable to thermal perturbations. Cool condensations could form, explaining features such as the dips in the X-ray light curves of V471 Tau, and the circumstellar material deduced to be present in several close binary systems.

The Sun is the only star whose X-ray emitting, strongly inhomogenous corona can be spatially resolved via direct observations. For other late type-stars it is known that coronae do exist, but the spatial distribution of their emission is largely unknown. However in the case of eclipsing binaries this spatial structure can be potentially deduced from the orbital modulation of the observed X-ray light curve. The best candidates for this kind of analysis are RS CVn binaries, the most active and luminous late-type X-ray coronal sources. These are detached binaries with periods typically between 0.5 and 20 days, in which one or both stars have evolved into subgiant or giant of spectral type G or K. For short orbital periods (< 14 days) the tidal forces lead to synchronization of the orbital and rotational periods, so these systems rotate rigidly.

The relation between photospheric and coronal active regions in late–type stars is studied from two different points of departure. First, I report on 5 years of ROSAT X–ray monitoring of the active young K–star AB Dor. I compare the X–ray data with 16 years of V–band brightness monitoring showing a 10–year decline between 1978 and 1989 and a subsequent rise phase. Quite differently, the X–ray flux of AB Dor (while exhibiting strong variability on time scales of minutes to weeks) reveals no pronounced long–term trend over the 5 years of the program. This supports the concept of a saturated corona. Second, I present rotationally modulated ROSAT X–ray light curves of three active stars (AB Dor, CF Tuc, YY Men) and compare them with contemporaneous Doppler images. I demonstrate that it is possible to explain the X–ray light curves by coronal emission regions that are spatially related with photospheric active regions. I discuss the concept of X–ray bright loops connecting the major star spot complexes.

Selected results relevant to coronal structuring in solar-like single stars based on combined radio and X-ray information are presented. Nonthermal radio and thermal soft X-ray emissions from coronal solar-like stars provide direct information on particle acceleration and coronal heating in the magnetically confined outer atmospheres. The structural relationship between the emission sources is mostly inferred from indirect arguments such as rotational modulation or gyroresonance emission. Direct VLBI provides evidence of resolved, extended coronae on solar-like stars.

The technique of back-projection Doppler tomography has been used to produce the first set of reconstructed images of chromospheric emission sources in the class of Algol-type binaries. The Hα emission sources in the Doppler tomograms include the gas stream, a sub-Keplerian accretion annulus, a quasi-stable accretion disk, and the chromosphere of the cool secondary star. The chromospheric images of several Algol secondaries (e.g., RS Vul, β Per, TX UMa, RW Tau) are consistent with that of the RS CVn binary V711 Tau.

Continued progress in the physical interpretation of solar and stellar observations, the increased speed of computing now available, and new highly efficient numerical procedures have all led to significant advances in stellar atmospheric modeling. This review attempts to summarize recent progress in the field and to describe the many interrelated physical processes that should be taken into account in next-generation modeling programs.

For the diagnostics of stellar surface structure, studies of “ordinary” line profiles may prove inadequate. However, hydrodynamic models may be constrained by second-order quantities, such as line asymmetries and wavelength shifts (and especially the differential behavior between lines of different excitation potential, ionization stage, and height of formation, as well as by the time dependence).

I have developed a new opacity-sampling version of my model atmosphere program called ATLAS12. It recognizes more than 1000 atomic and molecular species, each in up to 10 isotopic forms. It can treat all ions of the elements up through Zn and the first 5 ions of heavier elements up through Es. The elemental and isotopic abundances are treated as variables with depth. At present ATLAS12 reads atomic line data from Kurucz CD-ROM 1 and diatomic molecular data from Kurucz CD-ROM 15. For each line the wavelength, identification, lower energy level, gf, radiative, Stark, and van der Waals damping constants are packed into 16 bytes. At each wavelength point in a frequency integration the Voigt profiles of all the significant nearby lines are computed and summed. The H, He, and auto-ionizing lines are treated with the appropriate profile functions.

The fluxes predicted by ATLAS12 are not accurate in intermediate or narrow bandpass intervals because the sample size is too small. A special stripped version of the spectrum synthesis program SYNTHE is used to generate the surface flux for the converged model using the line data on CD-ROMs 1 and 15. That high resolution spectrum can be used directly or can be instrumentally broadened to compare to low resolution observations.

ATLAS12 can be used to produce improved models for Am and Ap stars. It should be very useful for investigating diffusion effects in atmospheres. It can be used to model exciting stars for H II regions with abundances consistent with those of the H II region. These programs and line files will be distributed on CD-ROMs.

Since its effective temperature, surface gravity and chemical composition are well established, the Sun has often been used for tests of model stellar photospheres. There are two important potential tests – firstly, comparisons of observed and calculated line spectra and, secondly, comparisons of observed and calculated fluxes. If solar model calculations can successfully reproduce both of these data sets, then it gives confidence in the quality of models of other cool stars. However, different observations of solar fluxes do not agree with one another with the precision required in testing solar models.

The atmospheres of M stars are dominated by a small number of very strong molecular compounds (H2O, TiO, H2, CO, VO). Most of the hydrogen is locked in molecular H2, most of the carbon in CO; and H2O, TiO and VO opacities define a pseudo-continuum covering the entire flux distribution of these stars. The optical “continuum” is due to TiO vibrational bands which are often used as temperature indicators for these stars. These may be the depth of the bands relative to the troughs in between them; or the depth of the VO bands; or of the atomic lines relative to the local “continuum”; or even the strength of the infrared water bands; all of these depend on the strength of the TiO bands and the amount of flux-redistribution to longer wavelengths exerted by them. Departures from LTE of the Ti I atom, and thus the concentration of the important TiO molecule, could, therefore, have severe and measurable consequences on the atmospheric structure and spectra of these stars.

We present the results of a long term research programme on the outer atmospheres of main-sequence dwarfs. Combining NLTE-radiation transfer calculations with high resolution spectroscopic observations have led to significant progress in understanding chromospheric physical properties and spectral signatures. We emphasize that in order to unravel the extremely complex physics of the outer atmosphere and its energy source, magnetic field and acoustic wave dissipation, one must isolate the influence of all stellar parameters.

Over the past decades various forms of the mixing length theory (MLT) have been used to describe convection in stellar atmospheres. Recent advances in turbulence theory now allow for major improvements in modelling thermal convection. We review several models for convection which have been derived from turbulence theory, and describe one of them, the “CM model”, in detail. The CM model has been used in several stellar evolution and helioseismology codes during the last four years and has now been applied to model atmospheres. An overwiew comparing stellar atmosphere models based on the CM formulation with its MLT predecessors indicates improvements on model atmospheres for A and F stars.

This IAU Symposium 176 on the topic “Stellar Surface Structure” provided a much needed opportunity for instrumentalists, observers, and theoreticians to present their exciting new results and to preview the even more exciting future for high-resolution imaging of stellar surfaces. I will attempt a critical summary of this symposium by calling attention to those areas that will likely be very productive in the future and by discussing some important topics that have not been addressed adequately to date, such as the roles of systematic and random errors, the true nature of the solar-stellar connection, and the physical processes responsible for stellar surface structure.

Lesser artists borrow, great artists steal. Igor Stravinsky

You never can tell. George Bernard Shaw