Automatic classification of eclipsing binaries light curves using neural networks
In this work we present a system for the automatic classification of thelight curves of eclipsing binaries. This system is based on aclassification scheme that aims to separate eclipsing binary systemsaccording to their geometrical configuration in a modified version ofthe traditional classification scheme. The classification is performedby a Bayesian ensemble of neural networks trained with Hipparcos data ofseven different categories including eccentric binary systems and twotypes of pulsating light curve morphologies.
|
Aus der Sektion Bedeckungsveranderliche.
Not Available
|
Einige interessante Bedeckungsveraenderliche.
Not Available
|
Studies of Intermediate-Mass Stellar Models Using Eclipsing Binaries
Evolutionary computations for intermediate-mass stars are analyzed usingobserved parameters for eclipsing SB2 binaries and theoreticalparameters based on evolutionary tracks. Modern observations cannot beused to distinguish between models with and without convectiveovershooting for stars in the vicinity of the main sequence.Statistically significant discrepancies between the observed andcomputed stellar parameters are associated with systematic errors inphotometric effective temperatures. After taking into account systematiceffects, the theoretical computations fit the observational datauniformly well throughout the entire mass interval studied. Empiricaland semiempirical (i.e., reduced to the ZAMS and with solar elementalabundances) formulas for the mass-luminosity, mass-effectivetemperature, and mass-radius relations are proposed.
|
Determination of the Ages of Close Binary Stars on the Main Sequence from Evolutionary Model Stars of Claret and Gimenez
A grid of isochrones, covering a wide range of stellar ages from thezero-age main sequence to 10 billion years, is calculated in the presentwork on the basis of the model stars of Claret and Gimenez withallowance for convective overshoot and mass loss by the components. Theages of 88 eclipsing variables on the main sequence from Andersen'scatalog and 100 chromospherically active stars from Strassmeier'scatalog are calculated with a description of the method of optimuminterpolation. Comparisons with age determinations by other authors aregiven and good agreement is established.
|
Detached double-lined eclipsing binaries as critical tests of stellar evolution. Age and metallicity determinations from the HR diagram
Detached, double-lined spectroscopic binaries that are also eclipsingprovide the most accurate determinations of stellar mass, radius,temperature and distance-independent luminosity for each of theirindividual components, and hence constitute a stringent test ofsingle-star stellar evolution theory. We compile a large sample of 60non-interacting, well-detached systems mostly with typical errorssmaller than 2% for mass and radius and smaller than 5% for effectivetemperature, and compare them with the properties predicted by stellarevolutionary tracks from a minimization method. To assess the systematicerrors introduced by a given set of tracks, we compare the resultsobtained using three widely-used independent sets of tracks, computedwith different physical ingredients (the Geneva, Padova and Granadamodels). We also test the hypothesis that the components of thesesystems are coeval and have the same metallicity, and compare thederived ages and metallicities with the ones obtained by fitting asingle isochrone to the system. Overall, there is a good agreement amongthe different determinations, and we provide a comprehensive discussionon the sub-sample of systems which either present problems or haveestimated metallicities. Although within the errors the published trackscan fit most of the systems, a large degeneracy between age andmetallicity remains. The power of the test is thus limited because themetallicities of most of the systems are unknown. The full version ofTable 6 is only available in the electronic form athttp://www.edpsciences.org
|
Rotational velocities of A-type stars in the northern hemisphere. II. Measurement of v sin i
This work is the second part of the set of measurements of v sin i forA-type stars, begun by Royer et al. (\cite{Ror_02a}). Spectra of 249 B8to F2-type stars brighter than V=7 have been collected at Observatoirede Haute-Provence (OHP). Fourier transforms of several line profiles inthe range 4200-4600 Å are used to derive v sin i from thefrequency of the first zero. Statistical analysis of the sampleindicates that measurement error mainly depends on v sin i and thisrelative error of the rotational velocity is found to be about 5% onaverage. The systematic shift with respect to standard values fromSlettebak et al. (\cite{Slk_75}), previously found in the first paper,is here confirmed. Comparisons with data from the literature agree withour findings: v sin i values from Slettebak et al. are underestimatedand the relation between both scales follows a linear law ensuremath vsin inew = 1.03 v sin iold+7.7. Finally, thesedata are combined with those from the previous paper (Royer et al.\cite{Ror_02a}), together with the catalogue of Abt & Morrell(\cite{AbtMol95}). The resulting sample includes some 2150 stars withhomogenized rotational velocities. Based on observations made atObservatoire de Haute Provence (CNRS), France. Tables \ref{results} and\ref{merging} are only available in electronic form at the CDS viaanonymous ftp to cdsarc.u-strasbg.fr (130.79.125.5) or viahttp://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/393/897
|
New results on the apsidal-motion test to stellar structure and evolution including the effects of dynamic tides
We revised the current status of the apsidal-motion test to stellarstructure and evolution. The observational sample was increased by about50% in comparison to previous studies. Classical and relativisticsystems were analyzed simultaneously and only systems with accurateabsolute dimensions were considered. New interior models incorporatingrecent opacity tables, stellar rotation, mass loss, and moderate coreovershooting were used as theoretical tools to compare the predictedwith the observed shifts of the position of the periastron. The stellarmodels were computed for the precise observed masses and the adoptedchemical compositions are consistent with the corresponding tables ofopacities to avoid the inherent problems of interpolation in mass and in(X, Z). The derived chemical composition for each individual system wasused to infer the primordial helium content as well as a law ofenrichment. The values found are in good agreement with those obtainedfrom various independent sources. For the first time, the effects ofdynamic tides are taken into account systematically to determine thecontribution of the tidal distortion to the predicted apsidal-motionrate. The deviations between the apsidal-motion rates resulting from theclassical formula and those determined by taking into account theeffects of dynamic tides are presented as a function of the level ofsynchronism. For systems close to synchronisation, dynamic tides causedeviations with respect to the classical apsidal-motion formula due tothe effects of the compressibility of the stellar fluid. For systemswith higher rotational angular velocities, additional deviations due toresonances arise when the forcing frequencies of the dynamic tides comeinto the range of the free oscillation modes of the component stars. Theresulting comparison shows a good agreement between the observed andtheoretical apsidal-motion rates. No systematic effects in the sensethat models are less mass concentrated than real stars and nocorrelations with the evolutionary status of the systems were detected.
|
Rotational velocities of A-type stars. I. Measurement of v sin i in the southern hemisphere
Within the scope of a Key Programme determining fundamental parametersof stars observed by HIPPARCOS, spectra of 525 B8 to F2-type starsbrighter than V=8 have been collected at ESO. Fourier transforms ofseveral line profiles in the range 4200-4500 Å are used to derivev sin i from the frequency of the first zero. Statistical analysis ofthe sample indicates that measurement error is a function of v sin i andthis relative error of the rotational velocity is found to be about 6%on average. The results obtained are compared with data from theliterature. There is a systematic shift from standard values from\citet{Slk_75}, which are 10 to 12% lower than our findings. Comparisonswith other independent v sin i values tend to prove that those fromSlettebak et al. are underestimated. This effect is attributed to thepresence of binaries in the standard sample of Slettebak et al., and tothe model atmosphere they used. Based on observations made at theEuropean Southern Observatory (ESO), La Silla, Chile, in the frameworkof the Key Programme 5-004-43K. Table 4 is only available in electronicform at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.125.5)or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/381/105
|
Age and Metallicity Estimates for Moderate-Mass Stars in Eclipsing Binaries
We estimate the ages and metallicities for the components of 43 binarysystems using a compilation of accurate observational data on eclipsingbinaries for which lines of both components are visible in theirspectra, together with two independent modern sets of stellar evolutionmodels computed for a wide range of masses and chemical abundances. Theuncertainties of the resulting values are computed, and their stabilityis demonstrated. The ages and metallicity are compared with thosederived in other studies using different methods, as well as withindependent estimates from photometric observations and observations ofclusters. These comparisons con firm the reliability of our ageestimates. The resulting metallicities depend significantly on thechoice of theoretical model. Comparison with independent estimatesfavors the estimates based on the evolutionary tracks of the Genevagroup.
|
Surface Brightness--Color Relation for Eclipsing Binaries with ubvy Photometry
The relation between visual surface brightness and (b-y)_0 color indexis constructed for eclipsing binaries with Hipparcos parallaxes andphotometry in the Stromgren four-color ubvy system. The relation isconsistent with the Popper relation derived for stars with angulardiameters obtained mainly from interferometry or from the lunaroccultation method. It is shown that the Lutz-Kelker correction tomagnitudes is in this case negligible.
|
Estimating the ages of eclipsing variable DM-stars on the basis of the evolutionary star models by Maeder and Meynet
A set of isochrones covering a wide range of star ages from5\cdot106 to 1010 yr was built on the basis of thestellar models by A. Maeder and G. Meynet with overshooting and massloss for Population I stars with abundances (X, Y, Z) = 0.70, 0.28,0.02. The isochrones were used to compute the ages of 88 eclipsingvariable stars from the catalog by Andersen which lie on the mainsequence. The influence of initial data errors on the rezultes wasinvestigated. The ages derived are in good agreement with the results ofother authors.
|
The mass dependence of the overshooting parameter determined from eclipsing binary data
High-precision data on absolute dimensions of evolved eclipsing binarieshave been used for a quantitative evaluation of the significance ofconvective overshoot in the stellar core. Eight detached double-linedeclipsing binaries with components close to or beyond the terminal agemain sequence (TAMS) and masses between 2 and 12Msolar havebeen compared with evolutionary models using different overshootingparameters. The results are robust and indicate a systematic increase ofthe amount of convective overshoot with the stellar mass. Suchdetermination constitutes a fundamental point for stellar structure andevolution theory.
|
Chemical composition of eclipsing binaries: a new approach to the helium-to-metal enrichment ratio
The chemical enrichment law Y(Z) is studied by using detacheddouble-lined eclipsing binaries with accurate absolute dimensions andeffective temperatures. A sample of 50 suitable systems was collectedfrom the literature, and their effective temperatures were carefullyre-determined. The chemical composition of each of the systems wasobtained by comparison with stellar evolutionary models, under theassumption that they should fit an isochrone to the observed propertiesof the components. Evolutionary models covering a wide grid in Z and Ywere adopted for our study. An algorithm was developed for searching thebest-fitting chemical composition (and the age) for the systems, basedon the minimization of a χ2 function. The errors (andbiases) of these parameters were estimated by means of Monte Carlosimulations, with special care put on the correlations existing betweenthe errors of both components. In order to check the physicalconsistency of the results, we compared our metallicity values withempirical determinations, obtaining excellent coherence. Theindependently derived Z and Y values yielded a determination of thechemical enrichment law via weighted linear least-squares fit. Our valueof the slope, ΔY/ΔZ=2.2+/-0.8, is in good agreement withrecent results, but it has a smaller formal error and it is free ofsystematic effects. Linear extrapolation of the enrichment law to zerometals leads to an estimation of the primordial helium abundance ofYp=0.225+/-0.013, possibly affected by systematics in theeffective temperature determination.
|
On the Variability of A3-F0 Luminosity Class III-V Stars
I investigate the Hipparcos Satellite photometry of A3-F0 stars ofluminosity classes III-V to learn about their variability and identify afew stars for which further study is desirable.
|
ICCD Speckle Observations of Binary Stars. XXIII. Measurements during 1982-1997 from Six Telescopes, with 14 New Orbits
We present 2017 observations of 1286 binary stars, observed by means ofspeckle interferometry using six telescopes over a 15 year period from1982 April to 1997 June. These measurements constitute the 23dinstallment in CHARA's speckle program at 2 to 4 m class telescopes andinclude the second major collection of measurements from the MountWilson 100 inch (2.5 m) Hooker Telescope. Orbital elements are alsopresented for 14 systems, seven of which have had no previouslypublished orbital analyses.
|
Radial velocities of HIPPARCOS southern B8-F2 type stars
Radial velocities have been determined for a sample of B8-F2 type starsobserved by the Hipparcos satellite. Observations were obtained withinthe framework of an ESO key-program. Radial velocities have beenmeasured using a cross-correlation method, the templates being a grid ofsynthetic spectra. The obtained precision depends on effectivetemperature and projected rotational velocity of the star as well as ona possible asymmetry of the correlation peak generally due to secondarycomponents. New spectroscopic binaries have been detected from theseasymmetries and the variability of the measured radial velocity.Simulations of binary and triple systems have been performed. Forbinaries our results have been compared with Hipparcos binary data.Adding the variable radial velocities, the minimum binary fraction hasbeen found 60% for physical systems. Radial velocities have beendetermined for 581 B8-F2 stars, 159 being new. Taking into accountpublished radial velocities, 39% south A-type stars with V magnitudelower than 7.5 have a radial velocity. Based on observations obtained atthe European Southern Observatory (ESO, La Silla, Chile) and on datafrom the ESA Hipparcos astrometry satellite.}\fnmsep \thanks{Tables 7, 8and 9 are only available in electronic form at the CDS via anonymous ftpto cdsarc.u-strasbg.fr (130.79.128.5) or viahttp://cdsweb.u-strasbg.fr/Abstract.html
|
Metallicity-dependent effective temperature determination for eclipsing binaries from synthetic UVBY Strömgren photometry
Strömgren synthetic photometry from an empirically calibrated gridof stellar atmosphere models has been used to derive the effectivetemperature of each component of double lined spectroscopic (SB2)eclipsing binaries. For this purpose, we have selected a sub-sample of20 SB2s for which (b-y), m_1, and c_1 individual indices are available.This new determination of effective temperature has been performed in ahomogeneous way for all these stars. As the effective temperaturedetermination is related to the assumed metallicity, we exploresimultaneous solutions in the ({{T_eff}},{[Fe/H]})-plane and present ourresults as confidence regions computed to match the observed values ofsurface gravity, (b-y), m_1, and c_1, taking into account interstellarreddening. These confidence regions show that previous estimates of{{T_eff}} are often too optimistic, and that {[Fe/H]} should not beneglected in such determinations. Comparisons with Ribas et al. (1998)using Hipparcos parallaxes are also presented for 8 binaries of ourworking sample, showing good agreement with the most reliableparallaxes. This point gives a significant weight to the validity of theBaSeL models for synthetic photometry applications.
|
Further critical tests of stellar evolution by means of double-lined eclipsing binaries
The most accurately measured stellar masses and radii come fromdetached, double-lined eclipsing binaries, as compiled by Andersen. Wepresent a detailed quantitative comparison of these fundamental datawith evolution models for single stars computed with our evolution code,both with and without the effects of enhanced mixing or overshootingbeyond the convective cores. We use the same prescription forovershooting that Schroder, Pols & Eggleton found to reproduce theproperties of zeta Aurigae binaries. For about 80 per cent of the 49binary systems in the sample, both sets of models provide a good fit toboth stars at a single age and metallicity within the observationaluncertainties. We discuss possible causes for the discrepancies in theother systems. For only one system, AI Hya, do the enhanced-mixingmodels provide a significantly better fit to the data. For two others(WX Cep and TZ For) the fit to the enhanced-mixing models is alsobetter. None of the other systems can individually distinguish betweenthe models with and without enhanced mixing. However, the number ofsystems in a post-main-sequence phase is in much better agreement withthe enhanced-mixing models. This test provides supportive evidence forextended mixing in main-sequence stars in the range 2-3Msolar.
|
ICCD Speckle Observations of Binary Stars. XVII. Measurements During 1993-1995 From the Mount Wilson 2.5-M Telescope.
Abstract image available at:http://adsabs.harvard.edu/cgi-bin/nph-bib_query?1997AJ....114.1639H&db_key=AST
|
MSC - a catalogue of physical multiple stars
The MSC catalogue contains data on 612 physical multiple stars ofmultiplicity 3 to 7 which are hierarchical with few exceptions. Orbitalperiods, angular separations and mass ratios are estimated for eachsub-system. Orbital elements are given when available. The catalogue canbe accessed through CDS (Strasbourg). Half of the systems are within 100pc from the Sun. The comparison of the periods of close and widesub-systems reveals that there is no preferred period ratio and allpossible combinations of periods are found. The distribution of thelogarithms of short periods is bimodal, probably due to observationalselection. In 82\% of triple stars the close sub-system is related tothe primary of a wide pair. However, the analysis of mass ratiodistribution gives some support to the idea that component masses areindependently selected from the Salpeter mass function. Orbits of wideand close sub-systems are not always coplanar, although thecorresponding orbital angular momentum vectors do show a weak tendencyof alignment. Some observational programs based on the MSC aresuggested. Tables 2 and 3 are only available in electronic form at theCDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp://cdsweb.u-strasbg.fr/Abstract.html
|
Photometric versus empirical surface gravities of eclipsing binaries.
Systematic differences in photometric stellar surface gravitydetermination are studied by means of the comparison with empiricalvalues derived from detached double-lined eclipsing binaries.Photometric gravities were computed using Moon & Dworetsky(1985MNRAS.217..305M) grids based on Kurucz (1979ApJS...40....1K)atmosphere models, and empirical gravities were taken from Andersen(1991A&ARv...3...91A). Individual Stroemgren colours and βindices of each component of the binary system have to be taken intoaccount to correctly analyze the observed differences. A compilation ofdata on a sample containing 30 detached double-lined eclipsing binarieswith accurate (=~1-2%) determination of mass and radius and availableuvbyHbeta_ photometric data is also presented. Correction ofthe differences in terms of T_eff_ and logg for the range11000K
|
Absolute dimensions of eclipsing binaries. XXI. V906 Scorpii: a triple system member of M 7.
We present an analysis based on new uvby light curves and spectroscopicdata of the detached triple-lined B-type eclipsing binary V906 Sco. Theuvby light curves are analysed with an extended version of theWilson-Devinney program. The spectroscopic CCD observations are analysedwith both the Sterne and the Lehmann-Filhes methods. We conclude fromthe combined analysis that the triple system V906 Sco is a member of theopen cluster M 7 and that its B-type eclipsing components are still onthe main sequence, at an age of (2.4+/-0.3)x10^8^yrs, already close tothe TAMS (especially the more massive one). The system is older than thetime for circularization of the orbit, and the small eccentricity isprobably caused by the third component, which, however, is unlikely tobe the main responsible for the apsidal motion, probably more influencedby tidal and rotational deformation. We determine absolute dimensions ofhigh precision (errors <2%): M_A_=3.25+/-0.07, R_A_=3.52+/-0.04,M_B_=3.38+/-0.07 and R_B_=4.52+/-0.04, in solar units. The system isdetached, with both components in synchronous rotation, and thereforerepresentative for normal stars.
|
The three-body system V 505 Sagittarii.
It is known that the period of the eclipsing binary V 505 Sgr changes,and that the changes can be explained - at least partly - by alight-time effect. The system is also known as a speckle binary. In thispaper parameters of the third body orbit are suggested in order toexplain both sets of data. Some discrepancies in the size of the orbitand radial velocity data remain. Next periastron passage should be at1998.9.
|
A Catalogue of Correlations Between Eclipsing Binaries and Other Categories of Double Stars
Among the 9110 stars in The Bright Star Catalogue, there are 225eclipsing or ellipsoidal variables. A search has been made for these incatalogues of spectroscopic binaries, visual double or multiple stars,speckle interferometry, occulation binaries, and galatic clusters. Themajority of the photometric binaries are also members of groups ofhigher multiplicity. The variables are in systems ranging from one to 91stars, five on the average. 199 are either spectroscopic binaries (SB)or stars with variable radial velocity, with orbital periods known for160. Photometric periods are lacking for 48 while SB periods areavailable for 23 of these. Observers with photoelectric equipment areencouraged to plan observations to test if the SB periods are consistentwith photometric data. Observers are likewise encouraged to examinethose stars for which the photometric and SB periods appear to beinconsistent. Parallaxes are available for 86 of the stars, 41 of themindicating distances nearer than 50 parsecs.
|
ICCD Speckle Observations of Binary Stars. XIII. Measurements During 1989- 1994 From the Cerro Tololo 4 M Telescope
Abstract image available at:http://adsabs.harvard.edu/cgi-bin/nph-bib_query?1996AJ....111..936H&db_key=AST
|
The Relation between Rotational Velocities and Spectral Peculiarities among A-Type Stars
Abstract image available at:http://adsabs.harvard.edu/cgi-bin/nph-bib_query?1995ApJS...99..135A&db_key=AST
|
Effective Temperature Bolometric Correction and Mass Calibration of O-F
Abstract image available at:http://adsabs.harvard.edu/cgi-bin/nph-bib_query?1994MNRAS.268..119B&db_key=AST
|
All-sky Stromgren photometry of speckle binary stars
All-sky Stromgren photometric observations were obtained for 303 specklebinaries. Most stars were in the range of V = 5-8. These data, whencombined with ratios of intensities from the CHARA speckle photometryprogram, will allow the determination of photometric indices for theindividual components of binary stars with separations as small as 0.05arcsec. These photometric indices will complement the stellar massesfrom the speckle interferometry observations to provide a much improvedmass-luminosity relationship.
|
