Search

We presented infrared photometry obtained with the IRAC camera on the Spitzer Space Telescope of a sample of 82 pre-main-sequence stars and brown dwarfs in the Taurus starforming region. We find a clear separation in some IRAC color-color diagrams between objects with and without disks. A few "transition" objects are noted, which correspond to systems in which the inner disk has been evacuated of small dust. Separating pure disk systems from objects with remnant protostellar envelopes is more difficult at IRAC wavelengths, especially for objects with infall at low rates and large angular momenta. Our results generally confirm the IRAC color classification scheme used in previous papers by Allen et al. and Megeath et al. to distinguish between protostars, T Tauri stars with disks, and young stars without (inner) disks. The observed IRAC colors are in good agreement with recent improved disk models, and in general accord with models for protostellar envelopes derived from analyzing a larger wavelength region. We also comment on a few Taurus objects of special interest. Our results should be useful for interpreting IRAC results in other, less well studied star-forming regions.

The Alfven wave-driven wind theory of Hartmann and MacGregor is applied to T Tauri variables, including modifications which permit the calculation of wind temperatures. It is shown that large wave fluxes generate low-temperature winds, which can radiate strongly in Balmer and other optical emission lines. If wave fluxes are restricted to be less than the stellar luminosity, mass loss rates are restricted to values less than or equal to 10 to the -8th solar mass per year. Although these mass loss rates are low in comparison to many previous estimates, it is shown that the wind models produce optical and ultraviolet emission roughly consistent with observations. The theory predicts large wave amplitudes, so that 'turbulent' velocities are generally comparable to, or larger than, local expansion velocities in the optical line-emitting regions, thus making the Sobolev approximation invalid. It is suggested that the discrepancy between the mass loss rates predicted here and previously estimated values is due to the inapplicability of the Sobolev approximation in T Tauri winds.

Recent radio observations have indicated that appreciable, continuing mass loss is occurring from regions of star formation. The conditions under which massive protostellar objects may exhibit strong, rotationally driven winds are investigated. For plausible rotational velocities and magnetic field strengths, ejection of about 10,000 solar masses a year at speeds of 10-100 km/s can be maintained for time scales of about 0.0001 yr; these values are insensitive to the protostellar luminosity. An evolutionary scenario is sketched which suggests that protostellar clouds may be expected to be rapidly rotating and to possess substantial magnetic fields when the clouds have contracted to radii of about 10 to the 14th cm. The centrifugally driven winds proposed present an alternative to fragmentation for providing the angular momentum loss required in the phase of rapid cloud contraction.

The wave-driven wind theory of Hartmann and MacGregor (1980) is extended to include effects due to non-radial divergence of the flow. Specifically, isothermal expansion within a flow tube whose cross-sectional area increases outward faster than the square of the radius near the stellar surface is considered. It is found that the qualitative conclusions of Hartmann and MacGregor concerning the physical properties of Alfven wave-driven winds are largely unaffected. In particular, mass fluxes of similar magnitude are obtained, and wave dissipation is still necessary to produce acceptably small terminal velocities. Increasingly divergent flow geometries generally lead to higher initial wind speeds and slightly lower terminal velocities. For some cases of extremely rapid flow tube divergence, steady supersonic wind solutions which extend to infinity with vanishing gas pressure cannot be obtained. In addition, departures from spherical symmetry can cause the relative Alfven wave amplitude delta-B/B to become approximately greater than 1 within several stellar radii of the base of the wind, suggesting that nonlinear processes may contribute to the wave dissipation required by the theory.

A survey is presented of chromospheric and transition region (TR) line emission in late-type stars obtained with the IUE satellite. A wide variation in TR emission (ions formed near T = 100,000 K) at a given position in the H-R diagram is found, contradicting theories that incorporate only acoustic energy generation and deposition. The behavior of the Hyades giants is thought to suggest that age does not uniquely determine emission levels. The data are consistent with the picture advanced by many others that rapid rotation appears to result in enhanced fluxes. Emission in TR lines is correlated with chromospheric emission; the range in observed surface fluxes from different stars increases with the temperature of line formation. It is noted that although emission measures of high temperature gas generally decline with decreasing stellar gravity, it is not possible to define a sharp division in the H-R diagram between stars with TR emitting regions and those without.

The short time-scale variability of singly ionized calcium chromospheric emission has been investigated in a few late-type stars. Emission-line variations with time scales of a few minutes to hours are seen in Alpha Tau (K5 III), Lambda And (G8 III-IV), and Epsilon Eri (K2 V). The existence of substantial chromospheric flux changes (10 to the 30th to 10 to the 32nd ergs) over short periods of time suggests that the calcium emission arises from a few small, coherent regions. Frequencies present in the data are discussed in the context of acoustic wave predictions and estimated acoustic cutoff frequencies for giants and dwarfs.

Ultraviolet emission spectra of several filaments in the Vela supernova remnant and the Cygnus Loop are compared with theoretical models of radiative shocks. Shock velocities range from 85 to 130 km/s. Little or no grain depletion is apparent. Several of the observed filaments deviate from the steady-flow models in the sense that much of the photoionization-recombination zone is absent. It is shown that resonance line scattering within the emitting region can account for the weakness of the C II 1335 and C IV 1550 lines. Evaporation fronts, which are basic to the McKee and Ostriker picture of supernova remnant evolution, do not contribute significantly to the observed emission. Some discrepancies in UV emission predicted by different sets of shock models are traced to differing atomic rates used in the calculations. A fluorescent emission line of H2 may be observable in some supernova remnants.

High-dispersion spectra taken with the IUE are used to examine the temperature stratification of the wind of the hybrid atmosphere star Alpha TrA (K4 II). C IV emission is shown to have a line width of 150-200 km/s, with Si III and C III exhibiting 100 km/s widths. It is suggested that the line widths reflect wind expansion, with C IV formed in an extended region flowing near the wind terminal velocity of 85 km/s. It is shown that the observed line widths can be accounted for by the stellar wind theory of Hartmann and MacGregor, in which the schematic dissipation of Alfven waves drives the outflow in addition to heating the wind. The observations of the warm expanding corona in Alpha TrA suggest a continuous progression between the high temperature low-mass-flux solar wind and the winds of luminous cool stars.

Simultaneous and quasi-simultaneous optical, UV and X-ray observations of the accreting degenerate dwarf stars AM Herculis, U Geminorum and SS Cygni are reported. The observations were obtained in March 1979, by instruments on board the Einstein Observatory and the IUE satellite, and optical properties were monitored on the ground. AM Her was found to be in a high state at the time of the observations, while SS Cyg and U Gem appeared to be in the optical low state. The presence of a strong UV excess is found in SS Cyg, U Gem and AM Her, which most likely originates from the boundary of the accretion disk in U Gem. The observed excess, with a blackbody component greater than 10 eV, is noted to be inconsistent with standard accretion disk and column models. It is suggested that nuclear burning at the surface of the white dwarf may be responsible for the excess UV flux in the three systems, with differences in spectral distributions resulting from different levels of magnetic field intensity and accretion rate.

During April, May, and July of 1978, HDE 226868, the optical counterpart of Cygnus X-1, was repeatedly observed in the ultraviolet with the IUE satellite. Some X-ray and infrared observations have been made during the same period. The general shape of the spectrum is that expected from a late O supergiant. Strong absorption features are apparent in the ultraviolet, some of which have been identified. The equivalent widths of the most prominent lines appear to be modulated with the orbital phase. This modulation is discussed in terms of the ionization contours calculated by Hatchett and McCray, for a binary X-ray source in the stellar wind of the companion.

The present study calculates the response of the outer atmospheres of cool low-gravity stars to the passage of the mechanical energy fluxes of solar magnitude in the form of acoustic waves and Alfven waves. It is shown that Alfven waves are efficient in generating outflow, and can account for the order of magnitude of observed mass loss in late-type luminous stars. However, unless these magnetic waves undergo some dissipation within several stellar radii of the surface, the predicted terminal velocities of the resulting stellar winds are far too high. Alfven wave dissipation should give rise to extended warm chromospheres in low-gravity late-type stars, a prediction which can be observationally tested.

Interstellar absorption lines have been observed toward 25 stars with the International Ultraviolet Explorer satellite. Results are presented for observations of interstellar lines of C III, C IV, N V, Si III, and Si IV. Strong lines of Si IV or C IV are seen toward 11 stars. The strengths of these lines are inconsistent with their formation in the same hot gas responsible for the interstellar O IV lines observed with the Copernicus ultraviolet spectrometer. Comparison with simple models of the ionization structures of H II regions around hot stars suggests that the observed column densities of Si III, Si IV, and C IV are in harmony with those expected in normal photoionized nebulae. The line of sight to HD 93250, and O3 V star near Eta Carinae, passes through highly ionized gas that shows features spanning 250 km/s in radial velocity. Stellar wind velocities are presented for 21 of the stars observed in this survey of interstellar absorption lines.

UV spectra (obtained by the IUE) of HD 77581, associated with the binary X-ray source Vela X-1, show a spectrum typical of an early B-type supergiant. However, the P Cygni profiles of strong resonance lines show substantial variations with orbital phase. These variations can be ascribed to the changing ionization state in the stellar wind caused by the X-ray emitting companion. X-ray and spectroscopic and photometric optical observations, simultaneous with the IUE observations, indicate behavior consistent with previous epochs.

The analysis of IUE observations of a filament on the western edge of the Cygnus Loop yields a shock velocity of about 130 km/sec, significantly higher than the velocity estimated from the optical spectrum. This velocity together with the high forbidden O III/H-beta ratio reported by Miller (1974) indicates that much of the hydrogen recombination region predicted by steady flow shock models is absent. This situation may occur because this part of the supernova remnant is in the shell-formation phase that accompanies the transition from the adiabatic to the snowplow phase or because the observed filament is a recently shocked cloud.

The results of a campaign of monitoring HZ Her with the IUE satellite are presented. The spectrum and light variation outside of eclipse are consistent with the Milgrom and Salpeter model for emission from the X-ray heated photosphere of the primary. Near eclipse there is UV flux in excess of the prediction of the heated photosphere model, which shows significant 35 day variations and is consistent with an origin in a precessing disk, as described by Gerend and Boynton. The spectrum shows the strong emission lines of N V and C IV, which generally vary in the sense of the continuum. The N V/C IV ratio is greater than 2 near orbital phase 0.5 and approaches unity near eclipse. It is suggested that these emission lines are optically thick and arise from both the heated photosphere and the accretion disk.

Ultraviolet spectra showing evidence for hybrid chromospheres and winds are found in the supergiant Alpha Aqr (G2 Ib) and Beta Aqr (G0 Ib). Characteristics of both solar-type transition regions and the cool chromospheric regions typical of later supergiants such as Lambda Vel (K5 Ib) are present. This suggests a gradual transition between activity and winds analogous to solar conditions and the cool, massive winds of luminous late-type stars.

The implications of recent ultraviolet observations of stellar transition-region lines for calculations of the Ca II and Mg II resonance lines are investigated. It is found that the adoption of high transition-region pressures for stars with active chromospheres, such as Lambda And and Alpha Aur, can be consistent with observed Ca II fluxes, contrary to the results obtained by Kelch et al. (1978) for Alpha Aur. Furthermore, the adoption of the high-pressure models removes a long-standing difficulty in the line profile calculations, since the deep central absorption present in earlier calculations is less pronounced or absent, in closer agreement with observations. The apparent contradiction between these models and the recent density diagnostic of Doschek et al. (1978) is also discussed.

Ultraviolet spectroscopic measurements of the M dwarf binary EQ Peg and the G8 V star Xi Boo A have been made with the International Ultraviolet Explorer satellite. High-temperature emission lines (N V, C IV, Si IV) are present with similar strengths in both stars. The surface fluxes of chromospheric and transition-region lines are enhanced relative to those of normal main-sequence stars and are comparable to those observed in solar active regions. If inhomogeneities are present, the surface fluxes may be markedly higher than the derived values. It is likely that the emission observed from EQ Peg was not the result of major flare activity; however, the quiescent radiative losses observed in the ultraviolet are comparable to the time-averaged optical flare losses. The similarity of the line emissions of active dwarfs of diverse spectral types suggests that the character of the transition region is independent of the stellar effective temperature.

The velocity perturbation associated with an outwardly propagating sound wave in a radiation-driven stellar wind gives rise to a periodic Doppler shifting of absorption lines formed in the flow. A linearized theory applicable to optically thin waves is used to show that the resulting fluctuation in the absorption-line force can cause the wave amplitude to grow. Detailed calculations of the acceleration due to a large number of lines indicate that significant amplification can occur throughout the high-velocity portion of winds in which the dominant force-producing lines have appreciable optical depths. In the particular case of the wind of Zeta Pup (O4f), it is found that the e-folding distance for wave growth is considerably shorter than the scale lengths over which the physical properties of the flow vary. A qualitative estimate of the rate at which mechanical energy due to nonlinear waves can be dissipated suggests that this mechanism may be important in heating the supersonic portion of winds of early-type stars.

The implications for convection in late type stars arising from the observations of BY Draconis type variability are examined. The possibility that the total luminosity of such a star is not constant is emphasized, and further observational tests to better define the variability are suggested. An alternative to standard spot models is explored in which the 'missing' flux from 'dark' spots is temporally redistributed; this model makes definite predictions about the correlation of mean light and color, and about the quiescent (i.e., unspotted) magnitudes of BY Draconis stars. The time scales of the long-period variability of these stars appear to require secular changes in convective energy transport. Consideration is given to the evidence for period changes in the optical light curves, and it is concluded that the reality of such changes is far less certain than previously claimed.

Ultraviolet spectra of six globular clusters (M15, M92, NGC 1851, NGC 6624, 47 Tuc, and NGC 6752) have been taken with the International Ultraviolet Explorer satellite in order to investigate the surface brightness distributions in the central regions. The ultraviolet emission indicates composite stellar spectra. Short-wavelength emission (1200-1900 A) arises from blue horizontal-branch stars; long-wavelength emission (2100-3000 A) is characteristic of late-type horizontal-branch and giant stars. The surface brightness distribution is more compact at short wavelengths than at long wavelengths, suggesting a segregation of horizontal-branch stars. A dense core appears to be present in X-ray-emitting clusters and absent in those clusters without X-ray sources. The X-ray source in NGC 6624 may have been detected at short wavelengths (1300-1900 A).

Ultraviolet observations of the binary X-ray source AM Herculis from the IUE satellite show strong emission lines of ions from O I to N V, probably originating in photoionized gas, and a continuum which is partially eclipsed in phase with the X-ray eclipse. The emission lines have broad (about 600 km/s) and sharp (about 80 km/s) components at different velocities, as has been seen in optical lines. The continuum is interpreted as two components, a blackbody (blackbody kT of 25-30 eV) which accounts for the X-ray emission below 0.5 keV and the eclipsed part of the UV continuum, and a component which is never eclipsed, whose spectrum is roughly proportional to the inverse of frequency. Strong ultraviolet emission due to optically thick cyclotron emission, which has been predicted theoretically, is not observed. This presents a severe difficulty for the theory of accretion onto the magnetic pole of a white dwarf.

Photographic B magnitudes of the four spot stars BY Dra (dM0e), CC Eri (K7 Ve), AU Mic (dM2.5e), and YZ CMi (dM4.5e) are reported which considerably extend the time baseline of the photoelectric data on these stars. The long-term variation of the stars in yearly mean light is investigated. It is found that both BY Dra and CC Eri exhibit long-term B-magnitude variations with amplitudes of about 0.3 to 0.4 magnitude and times between maximum and minimum light near 30 yr. No apparent variability is observed for AU Mic and YZ CMi. The possibility is discussed that the variability of BY Dra and CC Eri is connected with a starspot cycle, for which the period would be around 60 yr.

Copernicus UV observations of O VI in early-type stars are used to infer the existence of coronae or hot stellar winds. Attention is given to the optical doublet of O VI at wavelengths 3811 and 3834, which is seen in the spectra of some central stars in planetary nebulae, and taken to be evidence of a coronal wind. It is noted (from observations of NGC 6751) that hot gas in the coronal model produces radiation at energies above 54 eV, which can affect He II Zanstra temperatures. He(+) ionizing radiation in population I early-type stars is also taken as an indication of hot winds.