SIMULATIONS OF STELLAR RADIAL VELOCITY AND SPECTRAL LINE BISECTOR VARIATIONS: I. NONRADIAL PULSATIONSHatzes, Artie P.
doi: 10.1086/133805pmid: N/A
The radial velocity and spectral variations are calculated for low-amplitude nonradial pulsations in slowly rotating stars (v sin i = 3 km s-1). Sectoral modes (l = -m) in the range l = 1-8 are considered. An empirical equation relating the pulsational amplitude to the "observed" stellar radial velocity amplitude is derived and this indicates that precise radial velocity techniques should be able to detect low-amplitude (<1 km s-1) pulsations for sectoral modes as high as l = 6. The spectral variations manifest themselves via changes in the span and curvature of the line bisector which are maximized for m = 4. These spectral variations can be used to distinguish between radial velocity variations due to pulsations from those due to the reflex motion of the star caused by the presence of a low-mass companion. However, these measurements require data taken at high resolution (lambda/delta-lambad > 100,000) and high signal-to-noise ratios (S/N > 500).
HD 105262, A NEWLY-DISCOVERED HR 4049 STAR WITH A LARGE PROPER MOTIONAbt, Helmut A.
doi: 10.1086/133806pmid: N/A
HD 105262 has a spectrum like that of HR 4049 in having Balmer liens as in a Ib supergiant but very weak lines due to He I, Ca II, Si II, Ti II, fe I, Fe II, and Mg II. It has a published proper motion of 0.057" ± 0.011" yr-1. If the star had the luminosity of a Ib supergiant, its space motion would be several times the escape velocity from the Galaxy. HD 105262 does not have the spectral appearance of a horizontal-branch or post AGB star. The most likely explanation for the spectrum is the model proposed for HR 4049 by Lamers et al. (1986), Waelkens et al. (1987), and Lambert et al. (1988), namely a low-mass proto-planetary nebula star in which the narrow Balmer lines are formed in an extended atmosphere.
ON THE AUTOMATIC DETERMINATION OF LIGHT-CURVE PARAMETERS FOR CEPHEID VARIABLESStetson, Peter B.
doi: 10.1086/133808pmid: N/A
A computerized algorithm for the automatic detection of Cepheid variables and for the estimation of their periods, amplitudes, and mean magnitudes from sparse data sets is presented. It is intended to be suitable for use in such programs as the measurement of Cepheid distances to external galaxies, for example with the Hubble Space Telescope. The reliability of the algorithm is tested by application to new photometric reductions of pre-repair HST images of the nearby Sdm galaxy IC 4182, with comparison to published analyses of the same data (Saha et al. 1994, ApJ, 425, 14).
TEMPLATE K LIGHT CURVES FOR RR LYRAE STARSJones, Rodney V.; Carney, Bruce W.; Fulbright, Jon P.
doi: 10.1086/133809pmid: N/A
We discuss a technique to compute using templates developed by fitting a Fourier series to existing high-quality K light curves of field RR Lyraes. We find that a series of order 2 is sufficient to model the light curves of first-overtone RRc variables, but 4 different 6th-order templates are needed for the fundamental RRab stars due to changes in the light curves that appear to correlate with the B amplitude. Applying the appropriate template to single-phase observations yield estimated values whose deviation from the true is randomly distributed over phase, and is of the same order of magnitude as the observational uncertainty, as long as the ephemeris phase is accurate. The addition of a second point, separated by at least 0.2 in phase from the first, allows the use of template-shifting to remove deviations that may arise from uncertainties in the ephemeris phase, and template and scaling factor selection, with final systematic errors reduced to less than 0.03 mag. We find that the use of templates yield superior results to those derived using other techniques, which can produce values that show systematic deviations over phase.
ORBITAL PERIODS OF THE DWARF NOVAE AR AND, AM CAS, AND PY PERTaylor, Cynthia J.; Thorstensen, John R.
doi: 10.1086/133810pmid: N/A
We present time-resolved spectroscopy of the dwarf-novae AR And, AM Cas, and PY Per. From velocities of H-alpha emission lines, we determine orbital periods of 0.16302 ± 0.00032 d (=3.91 h) for AR And and 0.15480 ± 0.00017 d (=3.72 h) for PY Per. Our period determination for AR And resolves a daily cycle-count ambiguity in a previous determination by Shafter, Veal, and Robinson (1995). No orbital period for PY Per has previously been published. The orbital period of Am Cas, determined here for hte first time, is constrained over a 46-night baseline to one of several precise values near 0.165 d (=3.96 h), the best of which are 0.16490 ± 0.00001 d and 0.16551 ± 0.0001 d. For PY Per, we use a single direct CCD exposure to extend the magnitude sequence of Misselt (1996). A comparison with the Downes & Shara atlas (1993) shows that PY Per had V ~ 19.8 when the Downes & Shara image was obtained, which is much fainter than the minimum listed in the General Catalogue of Variable Stars (Kholopov 1987). Our averaged spectra of all appear typical for dwarf novae, except that the Fe II lambda-5169 in AR And is unusally strong and weak features at lambda-5230, lambda-5270, and labmda-5322 (which we attribute to Fe II) also appear.
THE CONSTANCY OF [ALPHA/FE] IN GLOBULAR CLUSTERS OF DIFFERING [FE/H] AND AGECarney, Bruce W.
doi: 10.1086/133811pmid: N/A
We review the literature to derive estimates for the abundances of the four "alpha" elements' abundances relative to iron [O/Fe], [Si/Fe], [Ca/Fe], and [Ti/Fe], as a function of [Fe/H] and age, in the globular clusters. Magnesium is under-studied, and is depleted in some metal-poor cluster giants, and hence is not useful for study. While [O/Fe] values within clusters vary, due to mixing of ON cycle material into their photospheres, consideration of only those stars with low sodium abundances, [Na/Fe] ≤ 0.00, shows that the unmixed stars in clusters have uniformly high [O/Fe] values, from [Fe/H] = -2.24 to -0.58. The "mean alpha" values from two of the other three elements, silicon and titanium, also do not appear to vary. [Ca/Fe] does appear to drop as [Fe/H] increases or age decreases, but this may indicate a problem with the use of neutral calcium lines. Thus, there does not appear to be any sign of Type Ia supernovae contributions to the globular clusters, despite wide ranges in [Fe/H] and age. If the three globular cluster classes, "old halo", "young halo", and "disk", are related to one another, if our relative cluster ages are correct, and if the Type Ia supernovae model is the correct explanation of the declining [alpha/Fe] ratios observed among field stars, then its timescale must be much longer than 109 years. If the timescale is that short, then at a minimum, it is clear that the "old halo" and "disk" globular clusters do not share a common chemical history, despite both showing prograde Galactic rotation. In that case, one of the classes presumably formed farther from the Galaxy and was later accreted by it.
ABUNDANCES OF SELECTED ELEMENTS IN FIVE OXYGEN-POOR STARS IN OMEGA CENTAURIZucker, Daniel; Wallerstein, George; Brown, Jeffery A.
doi: 10.1086/133812pmid: N/A
we have derived abundances or upper limits for Fe, O, Na, Al, Sc, and Eu in five oxygen-poor red giants in Omega Centauri, a globular cluster which displays a metallicity range of some 1.5 dex. All of the stars show large Al excesses and slight Sc excesses, and all but one reveal an excess of Na. We find either solar or slightly deficient Eu abundances in these stars, in agreement with other recent results. We have also used spectral synthesis to derive 12C/13C ratios which are similar to ratios obtianed for oxygen-rich stars in Omega Centauri and in other globular clusters.
A SUPERGIANT-DOMINATED STARBURST IN THE NUCLEUS OF NGC 4569Keel, William C.
doi: 10.1086/133813pmid: N/A
The Virgo spiral NGC 4569 has a very compact optical nucleus whose spectrum is highly composite, which strong Balmer absorption indicating that A stars dominate the optical light. Recent HST imaging shows that the nucleus is extremely compact in the mid-ultraviolet, so that either there must be a central AGN or an extraordinarily luminous and compact central star cluster. I present here IUE and ground-based spectrophotometry and high-resolution optical spectra, which are used to decompose the nucleus into possible stellar and AGN contributions. Accounting for the asymmetric contamination by Balmer emission, at a velocity different from the stellar absorption lines, removes the apparent discrepancy between radial velocities for different stellar components. The nuclear spectrum shows contributions from ordinary bulge light and from a young population. This young component has effective spectral type of mid-A, and the Balmer lines are so narrow that A-type supergiants are the most important contributors to the optical light. The young population is therefore much younger than the spectral type would suggest if interpreted as a main-sequence turnoff age, and the asociated OB stars might account for much of the UV flux. In this case, the nucleus contains an extraordinarily bright star cluster about as large as the core of 30 Doradus but more than 6 magnitudes brighter. While some less luminous supergiant-dominated star clusters have been found, current models of starburst evolution do not give a natural explanation for such a bunching in effective temperature. Evidence for an AGN is equivocal; M/L arguments show that only for the most extreme conditions of timing and burst duration could stars alone give a small enough value to satisfy the observational limit on velocity dispersion, and this condition would be eased if some of the UV radiation and the ionization of the nuclear gas can be attributed to an AGN. While, for very favorable conditions the Balmer absorption might be interpreted as coming from an unusually extensive and face-on accretion disk (rather than stars), the small linewidths and dominance of a single photospheric temperature place very stringent geometric and size constraints on this possibility.