BY Dra (MOVe+MOVe) and II Peg (K2IV-III) are well known noneclipsing spectroscopic binary systems showing the low-amplitude quasiperiodic photometric variability that is typical of spotted stars.

Since the discovery of their variability (Chugainov 1966, Eggen 1968) additional accurate photometry has been carried out (cf. Rodono 1982). On account of their highly variable light curves (LC), we have reanalyzed all the available observations and divided the original data into shorter time-interval sets, so that overlapping LCs with different shape could be separated. Additional LCs obtained at Catania Observatory till 198l were also included.

The light curve (LC) of the single-line spectroscopic binary system II Peg (K2-3, IV-III) generally shows an asymmetrical and highly variable shape (Fig.l, left). Therefore, single circular spot models can be successfully applied only to symmetric LCs, as the 1977 one by Vogt (1981). When asymmetrical or almost flat LCs develop, as appears to be the rule for II Peg, at leasts two-spot modeling is required (Bopp and Noah 1980, Dorren and Guinan 1982).

From classical single-spot models (Torres and Ferraz Mello 1973, Friedman and Gurtler 1975, Bopp and Evans 1973) we have developed a computer code including two separate circular spots of different size which are allowed to assume any relative location on the stellar surface. The model is symmetric with respect to the equator. The spots are assumed to radiate as back bodies with temperature ranging from 100 to 2000 degree lower than the temperature of the unspotted photosphere (4500°K).

Photometric observations in the VBLUW system (Lub, 1979) have been performed during 1980 and 1981 of 19 late G and early Ktype members of the Pleiades Cluster, in order to study their variability. All stars showed variations with amplitudes of 0.02 to 0.20 magn. in V. For 12 stars lightcurves were obtained which show periods that range from 0.24 to 1.22 days. The light curves are semi-regular and resemble those of BY Dra stars, although the periods are shorter.

Time series analysis of fluctuations of Ca II H and K chromospheric emission has provided us with much information concerning stellar activity. On all timescales, events which parallel solar behavior can be observed: activity cycles, on timescales of years; rotation of stars and evolution of active areas on timescales of days to weeks; flare-like phenomena on timescales as short as minutes.

We expect that the analogues of solar activity exist on other stars . By studying stellar counterparts to solar activity, we can hope to investigate the physical parameters which are thought to influence chromospheric and coronal activity. The stellar surfaces are usually spatially unresolvable; it is thus difficult to measure directly either small-scale surface inhomogeneities or the associated magnetic fields expected from spatially restricted areas.

On the Sun, however, areas with strong surface magnetic fields show intense chromospheric Ca II H and K emission (Babcock and Babcock 1955; Skumanich et al 1975). Although indirect, the Ca II H and K features are good indicators of stellar magnetic activity. A major advantage of the Ca II features is their accessibility to ground-based observatories. Long-term synoptic programs are in progress to monitor stellar chromospheric activity, and this paper will highlight ongoing work at Mt. Wilson. Monitoring variations of Ca II H and K chromospheric emission over different timescales can reveal different physical phenomena: (1) Long-term (years) variations corresponding to stellar activity cycles; (2) intermediate term (days-months) variations indicating rotation or evolution of stellar active areas; (3) short-term (minutes-hours) variations resulting from impulsive and flare-like phenomena.

The presence of an unseen companion to EV Lac with 2 to 4 Jupiter masses is likely. A period of ~45 years indicates the separation at the time of periastron is ~2 a.u.

It has long been noted that the characteristics of the intensity variations during a flare on UV Ceti stars depend on the observed spectral region. The intensity Increase being larger in the shorter wavelength region. Typical values for the ratioes between the relative intensity increases on EV Lac in the Johnson UBV-bands are òU/òB≈6 and òB/òV≈3. (Moffet 1974).

As for solar flares, one of the most physically revealing types of data for M-dwarf flares are high-resolution, time-resolved spectra. Due to the intrinsically faint nature of the M-dwarf stars, spectroscopic data has tended to be of low spectral (˜ 5 Ǻ) and temporal (˜ 5 min) resolution. However, with the development of image intensified spectrographs and fast, efficient digital detectors, the last several years have seen the successful acquisition of both high time and spectral resolution M-dwarf flare spectra. Recent programs have also been successfully conducted using the International Ultraviolet Explorer (IUE) satellite to obtain UV and EUV spectra of M-dwarf flares. These data reveal that dwarf M star flares are remarkably similar to solar flares in all aspects of their spectroscopic phenomenology.

Optical and ultraviolet spectroscopy can enable the assessment of the physical conditions characterizing a stellar flare atmosphere and thereby potentially elucidate the possible radiative and hydrodynamic transport mechanisms operative during stellar flares. In this review, I present illustrative examples of the spectroscopic diagnostic techniques that can be applied to the analysis of stellar flare spectroscopic data and the resulting inferences concerning stellar flare properties for M dwarf flare events.

We have carried out 45.4 hours of continuous spectroscopic and photoelectric B-band observations of AD Leo, DT Vir and YZ CMi. The spectroscopic resolution was about 1 A in wavelength and 5-10 min in time. 6 flares were recorded. Light curves and line profiles are to be presented in the Izv.Crim. Astrophys. Obs., Vol.69. In this paper only the main results are discussed.

Light curves of major stellar flares have been used to study the behavior of U-B, B-V, and V-R. The majority of the flux transmitted through these filters is continuum radiation, but U and B are affected by emission lines. The variability of Ha and H$ emission lines were monitored through narrow band filters. The timescales of emission line variability are considerably longer than those for the continuum, and the emission line flare peak occurs a few minutes after the continuum flare maximum. No variability in lines at a timescale of seconds is detected in our data.

Four large flare events – one on each of the dMe stars UV Cet, AT Mic, EV Lac and EQ Peg – have been witnessed during a total of 17½ hours of far-UV (λλ1150-1950) IUE exposures. Some observational characteristics of these four events are compared. Two showed strong enhancements of chromospheric and transition-region line strengths. The other two did not, even though their visible flares were intense (ΔU ~ 2 mag). The brightest UV flare spectrum (EQ Peg) is contrasted with that of the largest solar flare seen from ‘Skylab’.

A coordinated series of ground-based optical and IUE observations of BY Dra variables was undertaken to follow the spectral variation of these stars over one cycle. In the first series 20 LWR and 19 SWP trailed spectra were taken of AU Mic over a three day period 4-6 August 1980 .

In Figure 1 we show the mean integrated fluxes for the strong emission lines in the SWP spectra of AU Mic over the observed phase interval of 0.14 to 0.8 together with an approximate V light curve determined by the FES on IUE. From comparison of the emission line intensities and FES magnitudes in Figure 1 several points emerge.

Evidence for magnetic surface activity among cool stars of low gravity is discussed.

The history of stellar X-ray flare observations prior to EINSTEIN is reviewed. X-ray light curves as measured by the IPC are then presented for all time resolved flare events discovered as of July 1982 in the EINSTEIN data set. These light curves are analyzed in terms of solar-like loop models to derive densities, temperatures, loop lengths, magnetic field strength lower limits, etc. The failure of the model to adequately represent the observations in the case of the YZ CMi flares is discussed. The relationship of X-ray to optical emission and X-ray to UV emission is considered from both an observational and a theoretical viewpoint. It is concluded that the characterization of a flare by a single, time averaged ratio, Lx /Lopt , is not physically significant.

In reviewing radio studies of M-dwarf flare stars one is struck by the curious way in which the field developed. Indeed, that is twenty years old may be the greatest surprise, for if one imagines the largest solar flares occurring at distances comparable to the nearest stars the expected flux densities would be < 10 mJy. Yet, despite the fact that detection thresholds in 1963 were about two orders of magnitude higher than the expected value, Lovell et al. (1963) made extensive observations and reported the. detection of UV Ceti. This remarkable discovery was followed immediately by detections of V371 Ori (Slee et al., 1963) and EV Lac (Lovell et al., 1964). One might have thought that these unexpected discoveries would have spurred significant interest in this new field but they did not.

Observationally the study of solar flares has reached the stage where intensity-time distributions of emission over broad and resolved regions of the electromagnetic spectrum are obtained for spatially resolved parts of the flare. Polarization measurements add an important diagnostic tool in some wavebands but we shall not report on these here. In the optical band good ground based observations have been available for many years, whereas in the UV, soft X-ray and hard X-ray (> 5 keV) bands recent spacecraft have greatly extended the data base. Good high resolution maps are being made in the microwave region with the ground based VIA. We are now at the point where significant progress into understanding the flare problem has been made, and will continue to be made, during the current solar maximum. This coincides with the development of soft X-ray instruments sensitive enough to detect transient and quiescent emission from flare stars, particularly red dwarfs in the solar neighbourhood (e.g. Kahn et al,1979, Haisch et al, 1980) which previously had only been detected in the optical and radio wavebands.

The X-ray observations of solar flares by Japanese astronomy satellite Hinotori (ASTRO-A) are reviewed. Detailed results from the X-ray telescope and the soft X-ray spectrometers are given. The hard X-ray images in the 17-40 keV range show a wide variety of source structures. Examples of compact single source, stationary coronal source, and double sources are presented. High resolution spectra in the range 1.7-2.0A indicate strong turbulence and blue-shifted components in the beginning of flares. From the FeXXV and FeXXVI spectra two kinds of thermal plasma are shown to appear. The cooler component of Te=15-20xl06K, which increases in the impulsive phase of the hard X-ray burst, is suggested to be produced by the dissipation of the electron beams. The hotter component of Te=30-50xl06K and Ne=1011-1012 increases towards the flare maximum. This component is also responsible for emitting 17-40keV continuum, and emitted from a very compact source. The results are discussed in comparison with flare loop models. It is shown that evolutionary changes in the loop density realize various source structures in the hard X-ray under the unitary electron beam hypothesis.

XUV flare observations from SMM are discussed and a comparison is made with recent X-ray observations of stellar flares.

We report here on EUV observations of a sunspot observed by the Harvard instrument on Skylab. The observational data used here have been presented in a previous paper by Noyes et al. (1982), in which line identifications and intensities for the wavelength region 350 – 1350 A were given. Several electron density sensitive line ratios suggest a constant density, rather than constant pressure, emitting region, while temperature diagnostic line ratios of several ions yield temperatures below the temperatures expected in ionization equilibrium.

Braut and Noyes (1982,1983) have reported the detection of about 40 unidentified emission lines near 12µ in the solar spectrum. The strongest lines, at 811.578 cm-1 and 818.062 cm-1, respectively, appear as broad, shallow absorption lines, less than 3% deep, with central, emission reversals projecting 5-10% above the continuum. The emission lines strengthen at the limb and over spot penumbrae but seem to be absent over spot umbrae. The full width at half-intensity of the emission lines is about 5 km/sec, but the absorption widths are more than 10 times as broad. Over spot penumbrae, the Zeeman splitting of the emission lines is striking. The lines have the appearance of a Zeeman triplet; the central component is nearly absent at the center of the disk but is very strong near the limb where the field is viewed perpendicularly to the line of sight. The splitting over spot penumbrae is about 10 times the width of the central component, and is consistent with that of a spectral line with a Landé g-factor of unity in a magnetic field of 1500 gauss. Braut and Noyes (1982, 1983) point out that the 12 u lines are a potentially powerful tool for magnetic field measurements in stars. Further observational details will be found in their referenced papers.