WIKISKY.ORG
 Home Getting Started To Survive in the Universe News@Sky Astro Photo The Collection Forum Blog New! FAQ Press Login

# HD 108317

Contents

### Images

DSS Images   Other Images

### Related articles

 Galactic model parameters for field giants separated from field dwarfs by their 2MASS and V apparent magnitudesWe present a method which separates field dwarfs and field giants bytheir 2MASS and V apparent magnitudes. This method is based onspectroscopically selected standards and is hence reliable. We appliedit to stars in two fields, SA 54 and SA 82, and we estimated a full setof Galactic model parameters for giants including their total localspace density. Our results are in agreement with the ones given in therecent literature. Oxygen abundances in metal-poor subgiants as determined from [O I], O I and OH linesThe debate on the oxygen abundances of metal-poor stars has its originin contradictory results obtained using different abundance indicators.To achieve a better understanding of the problem we have acquired highquality spectra with the Ultraviolet and Visual Echelle Spectrograph atVLT, with a signal-to-noise of the order of 100 in the near ultravioletand 500 in the optical and near infrared wavelength range. Threedifferent oxygen abundance indicators, OH ultraviolet lines around 310.0nm, the [O i] line at 630.03 nm and the O i lines at 777.1-5 nm wereobserved in the spectra of 13 metal-poor subgiants with-3.0≤[Fe/H]≤-1.5. Oxygen abundances were obtained from theanalysis of these indicators which was carried out assuming localthermodynamic equilibrium and plane-parallel model atmospheres.Abundances derived from O i were corrected for departures from localthermodynamic equilibrium. Stellar parameters were computed usingT_eff-vs.-color calibrations based on the infrared flux method andBalmer line profiles, Hipparcos parallaxes and Fe II lines. [O/Fe]values derived from the forbidden line at 630.03 nm are consistent withan oxygen/iron ratio that varies linearly with [Fe/H] as[O/Fe]=-0.09(±0.08)[Fe/H]+0.36(±0.15). Values based on theO i triplet are on average 0.19±0.22 dex(s.d.) higher than thevalues based on the forbidden line while the agreement between OHultraviolet lines and the forbidden line is much better with a meandifference of the order of -0.09±0.25 dex(s.d.). In general, ourresults follow the same trend as previously published results with theexception of the ones based on OH ultraviolet lines. In that case ourresults lie below the values which gave rise to the oxygen abundancedebate for metal-poor stars. Li and Be depletion in metal-poor subgiantsA sample of metal-poor subgiants has been observed with the UVESspectrograph at the Very Large Telescope and abundances of Li and Behave been determined. Typical signal-to-noise per spectral bin valuesfor the co-added spectra are of the order of 500 for the ion{Li}{i} line(670.78 nm) and 100 for the ion{Be}{ii} doublet lines (313.04 nm). Thespectral analysis of the observations was carried out using the Uppsalasuite of codes and marcs (1D-LTE) model atmospheres with stellarparameters from photometry, parallaxes, isochrones and Fe ii lines.Abundance estimates of the light elements were corrected for departuresfrom local thermodynamic equilibrium in the line formation. Effectivetemperatures and Li abundances seem to be correlated and Be abundancescorrelate with [O/H]. Standard models predict Li and Be abundancesapproximately one order of magnitude lower than main-sequence valueswhich is in general agreement with the observations. On average, ourobserved depletions seem to be 0.1 dex smaller and between 0.2 and 0.4dex larger (depending on which reference is taken) than those predictedfor Li and Be, respectively. This is not surprising since the initial Liabundance, as derived from main-sequence stars on the Spite plateau, maybe systematically in error by 0.1 dex or more, and uncertainties in thespectrum normalisation and continuum drawing may affect our Beabundances systematically. Sulfur Abundances in Metal-Poor Stars Based on OAO-1.88m/HIDES SpectraThe LTE abundances of sulfur (S) of 21 metal-poor stars and one normalstar were explored in the metallicity range of -3 < [Fe/H] ≤ 0,based on the equivalent widths of the S I (1) 9212, 9237Å and S I(6) 8693, 9894Å lines measured on high-resolution spectra, whichwere observed by the OAO 1.88-m telescope equipped with HIDES. Our mainresults are: (1) The abundances derived from the S I (6) lines areconsistent with those from the S I (1) lines among our sample stars inthe range of [Fe/H] > -2 with an average difference of +0.03 ±0.05 dex, whereas a significant discrepancy is observed in the range of[Fe/H] ≤ -2. (2) The behavior of [S(6)/Fe], versus [Fe/H] of ourhalo sample stars exhibits a nearly flat trend with an average of +0.62± 0.09 dex in the range of -3 < [Fe/H] < -1.25, and shows adistribution around +0.29 dex in -1.25 ≤ [Fe/H] ≤ -0.7. Oursample stars with -1.25 ≤ [Fe/H] ≤ -0.5 follow an increasingtrend with decreasing [Fe/H]. The behavior of [S(1)/Fe] of our samplestars also shows essentially the same trend as [S(6)/Fe], though it isquantitatively different. (3) The S behavior in the range of -3 <[Fe/H] ≤ 0 inferred from the abundances of multiplets 6 and 1 arequalitatively consistent with each other, and may be represented by acombination of a nearly flat trend and a linearly increasing trend withdecreasing [Fe/H]. A transition of the trend is likely to occur at[Fe/H]  -1.5 dex. Estimation of Carbon Abundances in Metal-Poor Stars. I. Application to the Strong G-Band Stars of Beers, Preston, and ShectmanWe develop and test a method for the estimation of metallicities([Fe/H]) and carbon abundance ratios ([C/Fe]) for carbon-enhancedmetal-poor (CEMP) stars based on the application of artificial neuralnetworks, regressions, and synthesis models to medium-resolution (1-2Å) spectra and J-K colors. We calibrate this method by comparisonwith metallicities and carbon abundance determinations for 118 starswith available high-resolution analyses reported in the recentliterature. The neural network and regression approaches make use of apreviously defined set of line-strength indices quantifying the strengthof the Ca II K line and the CH G band, in conjunction with J-K colorsfrom the Two Micron All Sky Survey Point Source Catalog. The use ofnear-IR colors, as opposed to broadband B-V colors, is required becauseof the potentially large affect of strong molecular carbon bands onbluer color indices. We also explore the practicality of obtainingestimates of carbon abundances for metal-poor stars from the spectralinformation alone, i.e., without the additional information provided byphotometry, as many future samples of CEMP stars may lack such data. Wefind that although photometric information is required for theestimation of [Fe/H], it provides little improvement in our derivedestimates of [C/Fe], and hence, estimates of carbon-to-iron ratios basedsolely on line indices appear sufficiently accurate for most purposes.Although we find that the spectral synthesis approach yields the mostaccurate estimates of [C/Fe], in particular for the stars with thestrongest molecular bands, it is only marginally better than is obtainedfrom the line index approaches. Using these methods we are able toreproduce the previously measured [Fe/H] and [C/Fe] determinations withan accuracy of ~0.25 dex for stars in the metallicity interval-5.5<=[Fe/H]<=-1.0 and with 0.2<=(J-K)0<=0.8. Athigher metallicity, the Ca II K line begins to saturate, especially forthe cool stars in our program, and hence, this approach is not useful insome cases. As a first application, we estimate the abundances of [Fe/H]and [C/Fe] for the 56 stars identified as possibly carbon-rich, relativeto stars of similar metal abundance, in the sample of strong G-band''stars discussed by Beers, Preston, and Shectman. The lithium content of the Galactic Halo starsThanks to the accurate determination of the baryon density of theuniverse by the recent cosmic microwave background experiments, updatedpredictions of the standard model of Big Bang nucleosynthesis now yieldthe initial abundance of the primordial light elements withunprecedented precision. In the case of ^7Li, the CMB+SBBN value issignificantly higher than the generally reported abundances for Pop IIstars along the so-called Spite plateau. In view of the crucialimportance of this disagreement, which has cosmological, galactic andstellar implications, we decided to tackle the most critical issues ofthe problem by revisiting a large sample of literature Li data in halostars that we assembled following some strict selection criteria on thequality of the original analyses. In the first part of the paper wefocus on the systematic uncertainties affecting the determination of theLi abundances, one of our main goal being to look for the "highestobservational accuracy achievable" for one of the largest sets of Liabundances ever assembled. We explore in great detail the temperaturescale issue with a special emphasis on reddening. We derive four sets ofeffective temperatures by applying the same colour {T}_eff calibrationbut making four different assumptions about reddening and determine theLTE lithium values for each of them. We compute the NLTE corrections andapply them to the LTE lithium abundances. We then focus on our "best"(i.e. most consistent) set of temperatures in order to discuss theinferred mean Li value and dispersion in several {T}_eff and metallicityintervals. The resulting mean Li values along the plateau for [Fe/H]≤ 1.5 are A(Li)_NLTE = 2.214±0.093 and 2.224±0.075when the lowest effective temperature considered is taken equal to 5700K and 6000 K respectively. This is a factor of 2.48 to 2.81 (dependingon the adopted SBBN model and on the effective temperature range chosento delimit the plateau) lower than the CMB+SBBN determination. We findno evidence of intrinsic dispersion. Assuming the correctness of theCMB+SBBN prediction, we are then left with the conclusion that the Liabundance along the plateau is not the pristine one, but that halo starshave undergone surface depletion during their evolution. In the secondpart of the paper we further dissect our sample in search of newconstraints on Li depletion in halo stars. By means of the Hipparcosparallaxes, we derive the evolutionary status of each of our samplestars, and re-discuss our derived Li abundances. A very surprisingresult emerges for the first time from this examination. Namely, themean Li value as well as the dispersion appear to be lower (althoughfully compatible within the errors) for the dwarfs than for the turnoffand subgiant stars. For our most homogeneous dwarfs-only sample with[Fe/H] ≤ 1.5, the mean Li abundances are A(L)_NLTE = 2.177±0.071 and 2.215±0.074 when the lowest effective temperatureconsidered is taken equal to 5700 K and 6000 K respectively. This is afactor of 2.52 to 3.06 (depending on the selected range in {T}_eff forthe plateau and on the SBBN predictions we compare to) lower than theCMB+SBBN primordial value. Instead, for the post-main sequence stars thecorresponding values are 2.260±0.1 and 2.235±0.077, whichcorrespond to a depletion factor of 2.28 to 2.52. These results,together with the finding that all the stars with Li abnormalities(strong deficiency or high content) lie on or originate from the hotside of the plateau, lead us to suggest that the most massive of thehalo stars have had a slightly different Li history than their lessmassive contemporaries. In turn, this puts strong new constraints on thepossible depletion mechanisms and reinforces Li as a stellartomographer. Chemical abundances in 43 metal-poor starsWe have derived abundances of O, Na, Mg, Al, Si, Ca, Sc, Ti, V, Cr, Fe,Ni, and Ba for 43 metal-poor field stars in the solar neighbourhood,most of them subgiants or turn-off-point stars with iron abundances[Fe/H] ranging from -0.4 to -3.0. About half of this sample has not beenspectroscopically analysed in detail before. Effective temperatures wereestimated from uvby photometry, and surface gravities primarily fromHipparcos parallaxes. The analysis is differential relative to the Sun,and was carried out with plane-parallel MARCS models. Various sources oferror are discussed and found to contribute a total error of about0.1-0.2 dex for most elements, while relative abundances, such as[Ca/Fe], are most probably more accurate. For the oxygen abundances,determined in an NLTE analysis of the 7774 Å triplet lines, theerrors may be somewhat larger. We made a detailed comparison withsimilar studies and traced the reasons for the, in most cases,relatively small differences. Among the results we find that [O/Fe]possibly increases beyond [Fe/H] = -1.0, though considerably less sothan in results obtained by others from abundances based on OH lines. Wedid not trace any tendency toward strong overionization of iron, andfind the excesses, relative to Fe and the Sun, of the α elementsMg, Si, and Ca to be smaller than those of O. We discuss someindications that also the abundances of different α elementsrelative to Fe vary and the possibility that some of the scatter aroundthe trends in abundances relative to iron may be real. This may supportthe idea that the formation of Halo stars occurred in smaller systemswith different star formation rates. We verify the finding by Gratton etal. (2003b, A&A, 406, 131) that stars that do not participate in therotation of the galactic disk show a lower mean and larger spread in [α/Fe] than stars participating in the general rotation. The latterstars also seem to show some correlation between [ α/Fe] androtation speed. We trace some stars with peculiar abundances, amongthese two Ba stars, HD 17072 and HD196944, the second already known to be rich in s elements.Finally we advocate that a spectroscopic study of a larger sample ofhalo stars with well-defined selection criteria is very important, inorder to add to the very considerable efforts that various groups havealready made. The Indo-US Library of Coudé Feed Stellar SpectraWe have obtained spectra for 1273 stars using the 0.9 m coudéfeed telescope at Kitt Peak National Observatory. This telescope feedsthe coudé spectrograph of the 2.1 m telescope. The spectra havebeen obtained with the no. 5 camera of the coudé spectrograph anda Loral 3K×1K CCD. Two gratings have been used to provide spectralcoverage from 3460 to 9464 Å, at a resolution of ~1 Å FWHMand at an original dispersion of 0.44 Å pixel-1. For885 stars we have complete spectra over the entire 3460 to 9464 Åwavelength region (neglecting small gaps of less than 50 Å), andpartial spectral coverage for the remaining stars. The 1273 stars havebeen selected to provide broad coverage of the atmospheric parametersTeff, logg, and [Fe/H], as well as spectral type. The goal ofthe project is to provide a comprehensive library of stellar spectra foruse in the automated classification of stellar and galaxy spectra and ingalaxy population synthesis. In this paper we discuss thecharacteristics of the spectral library, viz., details of theobservations, data reduction procedures, and selection of stars. We alsopresent a few illustrations of the quality and information available inthe spectra. The first version of the complete spectral library is nowpublicly available from the National Optical Astronomy Observatory(NOAO) via ftp and http. The Rise of the s-Process in the GalaxyFrom newly obtained high-resolution, high signal-to-noise ratio spectrathe abundances of the elements La and Eu have been determined over thestellar metallicity range -3<[Fe/H]<+0.3 in 159 giant and dwarfstars. Lanthanum is predominantly made by the s-process in the solarsystem, while Eu owes most of its solar system abundance to ther-process. The changing ratio of these elements in stars over a widemetallicity range traces the changing contributions of these twoprocesses to the Galactic abundance mix. Large s-process abundances canbe the result of mass transfer from very evolved stars, so to identifythese cases we also report carbon abundances in our metal-poor stars.Results indicate that the s-process may be active as early as[Fe/H]=-2.6, although we also find that some stars as metal-rich as[Fe/H]=-1 show no strong indication of s-process enrichment. There is asignificant spread in the level of s-process enrichment even at solarmetallicity. Stellar Chemical Signatures and Hierarchical Galaxy FormationTo compare the chemistries of stars in the Milky Way dwarf spheroidal(dSph) satellite galaxies with stars in the Galaxy, we have compiled alarge sample of Galactic stellar abundances from the literature. Whenkinematic information is available, we have assigned the stars tostandard Galactic components through Bayesian classification based onGaussian velocity ellipsoids. As found in previous studies, the[α/Fe] ratios of most stars in the dSph galaxies are generallylower than similar metallicity Galactic stars in this extended sample.Our kinematically selected stars confirm this for the Galactic halo,thin-disk, and thick-disk components. There is marginal overlap in thelow [α/Fe] ratios between dSph stars and Galactic halo stars onextreme retrograde orbits (V<-420 km s-1), but this is notsupported by other element ratios. Other element ratios compared in thispaper include r- and s-process abundances, where we find a significantoffset in the [Y/Fe] ratios, which results in a large overabundance in[Ba/Y] in most dSph stars compared with Galactic stars. Thus, thechemical signatures of most of the dSph stars are distinct from thestars in each of the kinematic components of the Galaxy. This resultrules out continuous merging of low-mass galaxies similar to these dSphsatellites during the formation of the Galaxy. However, we do not ruleout very early merging of low-mass dwarf galaxies, since up to one-halfof the most metal-poor stars ([Fe/H]<=-1.8) have chemistries that arein fair agreement with Galactic halo stars. We also do not rule outmerging with higher mass galaxies, although we note that the LMC and theremnants of the Sgr dwarf galaxy are also chemically distinct from themajority of the Galactic halo stars. Formation of the Galaxy's thickdisk by heating of an old thin disk during a merger is also not ruledout; however, the Galaxy's thick disk itself cannot be comprised of theremnants from a low-mass (dSph) dwarf galaxy, nor of a high-mass dwarfgalaxy like the LMC or Sgr, because of differences in chemistry.The new and independent environments offered by the dSph galaxies alsoallow us to examine fundamental assumptions related to thenucleosynthesis of the elements. The metal-poor stars ([Fe/H]<=-1.8)in the dSph galaxies appear to have lower [Ca/Fe] and [Ti/Fe] than[Mg/Fe] ratios, unlike similar metallicity stars in the Galaxy.Predictions from the α-process (α-rich freeze-out) would beconsistent with this result if there have been a lack of hypernovae indSph galaxies. The α-process could also be responsible for thevery low Y abundances in the metal-poor stars in dSph's; since [La/Eu](and possibly [Ba/Eu]) are consistent with pure r-process results, thelow [Y/Eu] suggests a separate r-process site for this light(first-peak) r-process element. We also discuss SNe II rates and yieldsas other alternatives, however. In stars with higher metallicities([Fe/H]>=-1.8), contributions from the s-process are expected; [(Y,La, and Ba)/Eu] all rise as expected, and yet [Ba/Y] is still muchhigher in the dSph stars than similar metallicity Galactic stars. Thisresult is consistent with s-process contributions from lower metallicityAGB stars in dSph galaxies, and is in good agreement with the slowerchemical evolution expected in the low-mass dSph galaxies relative tothe Galaxy, such that the build-up of metals occurs over much longertimescales. Future investigations of nucleosynthetic constraints (aswell as galaxy formation and evolution) will require an examination ofmany stars within individual dwarf galaxies.Finally, the Na-Ni trend reported in 1997 by Nissen & Schuster isconfirmed in Galactic halo stars, but we discuss this in terms of thegeneral nucleosynthesis of neutron-rich elements. We do not confirm thatthe Na-Ni trend is related to the accretion of dSph galaxies in theGalactic halo. Empirically Constrained Color-Temperature Relations. II. uvbyA new grid of theoretical color indices for the Strömgren uvbyphotometric system has been derived from MARCS model atmospheres and SSGsynthetic spectra for cool dwarf and giant stars having-3.0<=[Fe/H]<=+0.5 and 3000<=Teff<=8000 K. Atwarmer temperatures (i.e., 8000-2.0. To overcome thisproblem, the theoretical indices at intermediate and high metallicitieshave been corrected using a set of color calibrations based on fieldstars having well-determined distances from Hipparcos, accurateTeff estimates from the infrared flux method, andspectroscopic [Fe/H] values. In contrast with Paper I, star clustersplayed only a minor role in this analysis in that they provided asupplementary constraint on the color corrections for cool dwarf starswith Teff<=5500 K. They were mainly used to test thecolor-Teff relations and, encouragingly, isochrones thatemploy the transformations derived in this study are able to reproducethe observed CMDs (involving u-v, v-b, and b-y colors) for a number ofopen and globular clusters (including M67, the Hyades, and 47 Tuc)rather well. Moreover, our interpretations of such data are verysimilar, if not identical, with those given in Paper I from aconsideration of BV(RI)C observations for the sameclusters-which provides a compelling argument in support of thecolor-Teff relations that are reported in both studies. Inthe present investigation, we have also analyzed the observedStrömgren photometry for the classic Population II subdwarfs,compared our final'' (b-y)-Teff relationship with thosederived empirically in a number of recent studies and examined in somedetail the dependence of the m1 index on [Fe/H].Based, in part, on observations made with the Nordic Optical Telescope,operated jointly on the island of La Palma by Denmark, Finland, Iceland,Norway, and Sweden, in the Spanish Observatorio del Roque de losMuchachos of the Instituto de Astrofisica de Canarias.Based, in part, on observations obtained with the Danish 1.54 mtelescope at the European Southern Observatory, La Silla, Chile. The Geneva-Copenhagen survey of the Solar neighbourhood. Ages, metallicities, and kinematic properties of 14 000 F and G dwarfsWe present and discuss new determinations of metallicity, rotation, age,kinematics, and Galactic orbits for a complete, magnitude-limited, andkinematically unbiased sample of 16 682 nearby F and G dwarf stars. Our63 000 new, accurate radial-velocity observations for nearly 13 500stars allow identification of most of the binary stars in the sampleand, together with published uvbyβ photometry, Hipparcosparallaxes, Tycho-2 proper motions, and a few earlier radial velocities,complete the kinematic information for 14 139 stars. These high-qualityvelocity data are supplemented by effective temperatures andmetallicities newly derived from recent and/or revised calibrations. Theremaining stars either lack Hipparcos data or have fast rotation. Amajor effort has been devoted to the determination of new isochrone agesfor all stars for which this is possible. Particular attention has beengiven to a realistic treatment of statistical biases and errorestimates, as standard techniques tend to underestimate these effectsand introduce spurious features in the age distributions. Our ages agreewell with those by Edvardsson et al. (\cite{edv93}), despite severalastrophysical and computational improvements since then. We demonstrate,however, how strong observational and theoretical biases cause thedistribution of the observed ages to be very different from that of thetrue age distribution of the sample. Among the many basic relations ofthe Galactic disk that can be reinvestigated from the data presentedhere, we revisit the metallicity distribution of the G dwarfs and theage-metallicity, age-velocity, and metallicity-velocity relations of theSolar neighbourhood. Our first results confirm the lack of metal-poor Gdwarfs relative to closed-box model predictions (the G dwarfproblem''), the existence of radial metallicity gradients in the disk,the small change in mean metallicity of the thin disk since itsformation and the substantial scatter in metallicity at all ages, andthe continuing kinematic heating of the thin disk with an efficiencyconsistent with that expected for a combination of spiral arms and giantmolecular clouds. Distinct features in the distribution of the Vcomponent of the space motion are extended in age and metallicity,corresponding to the effects of stochastic spiral waves rather thanclassical moving groups, and may complicate the identification ofthick-disk stars from kinematic criteria. More advanced analyses of thisrich material will require careful simulations of the selection criteriafor the sample and the distribution of observational errors.Based on observations made with the Danish 1.5-m telescope at ESO, LaSilla, Chile, and with the Swiss 1-m telescope at Observatoire deHaute-Provence, France.Complete Tables 1 and 2 are only available in electronic form at the CDSvia anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/418/989 Non-LTE Analysis of the Sodium Abundance of Metal-Poor Stars in the Galactic Disk and HaloWe performed an extensive non-LTE analysis of the neutral sodium linesof Na I 5683/5688, 5890/5896, 6154/6161, and 8183/8195 in disk/halostars of types F-K covering a wide metallicity range (-4 <≈[Fe/H] <≈ +0.4), using our own data as well as data collectedfrom the literature. For comparatively metal-rich disk stars (-1<≈ [Fe/H] <≈ +0.4) where the weaker 6154/6161 linesare the best abundance indicators, we confirmed [Na/Fe]  0 with an"upturn" (i.e., a shallow/broad dip around -0.5 <≈ [Fe/H]<≈ 0) as already reported in previous studies. For themetal-deficient halo stars, where the much stronger 5890/5896 or8183/8195 lines subject to considerable (negative) non-LTE correctionsamounting to 0.5 dex have to be used, our analysis suggests mildly"subsolar" [Na/Fe] values down to  -0.4 (with a somewhat largescatter of  ± 0.2 dex) on the average at the typical halometallicity of [Fe/H]  -2, followed by a rise again to a near-solarratio of [Na/Fe]  0 at the very metal-poor regime [Fe/H]  -3to -4. These results are discussed in comparison with the previousobservational studies along with the theoretical predictions from theavailable chemical evolution models. Oxygen Abundances in Metal-poor StarsWe present oxygen abundances derived from both the permitted andforbidden oxygen lines for 55 subgiants and giants with [Fe/H] valuesbetween -2.7 and solar with the goal of understanding the discrepancy inthe derived abundances. A first attempt, using Teff valuesfrom photometric calibrations and surface gravities from luminositiesobtained agreement between the indicators for turn-off stars, but thedisagreement was large for evolved stars. We find that the difference inthe oxygen abundances derived from the permitted and forbidden lines ismost strongly affected by Teff, and we derive a newTeff scale based on forcing the two sets of lines to give thesame oxygen abundances. These new parameters, however, do not agree withother observables, such as theoretical isochrones or Balmer-line profilebased Teff determinations. Our analysis finds thatone-dimensional, LTE analyses (with published non-LTE corrections forthe permitted lines) cannot fully resolve the disagreement in the twoindicators without adopting a temperature scale that is incompatiblewith other temperature indicators. We also find no evidence ofcircumstellar emission in the forbidden lines, removing such emission asa possible cause for the discrepancy. Spectroscopic Binaries, Velocity Jitter, and Rotation in Field Metal-poor Red Giant and Red Horizontal-Branch StarsWe summarize 2007 radial velocity measurements of 91 metal-poor fieldred giants. Excluding binary systems with orbital solutions, ourcoverage averages 13.7 yr per star, with a maximum of 18.0 yr. We reportfour significant findings. (1) Sixteen stars are found to bespectroscopic binaries, and we present orbital solutions for 14 of them.The spectroscopic binary frequency of the metal-poor red giants, with[Fe/H]<=-1.4, for periods less than 6000 days, is 16%+/-4%, which isnot significantly different from that of comparable-metallicity fielddwarfs, 17%+/-2%. The two CH stars in our program, BD -1°2582 and HD135148, are both spectroscopic binaries. (2) Velocity jitter is presentamong about 40% of the giants with MV<=-1.4. The twobest-observed cases, HD 3008 and BD +22°2411, showpseudoperiodicities of 172 and 186 days, longer than any knownlong-period variable in metal-poor globular clusters. Photometricvariability seen in HD 3008 and three other stars showing velocityjitter hints that starspots are the cause. However, the phasing of thevelocity data with the photometry data from Hipparcos is not consistentwith a simple starspot model for HD 3008. We argue against orbitalmotion effects and radial pulsation, so rotational modulation remainsthe best explanation. The implied rotational velocities for HD 3008 andBD +22°2411, both with MV<=-1.4 and R~50Rsolar, exceed 12 km s-1. (3) Including HD 3008and BD +22°2411, we have found signs of significant excess linebroadening in eight of the 17 red giants with MV<=-1.4,which we interpret as rotation. In three cases, BD +30°2034, CD-37°14010, and HD 218732, the rotation is probably induced by tidallocking between axial rotation and the observed orbital motion with astellar companion. But this cannot explain the other five stars in oursample that display signs of significant rotation. This high frequencyof elevated rotational velocities does not appear to be caused bystellar mass transfer or mergers: there are too few main-sequencebinaries with short enough periods. We also note that the lack of anynoticeable increase in mean rotation at the magnitude level of the redgiant branch luminosity function bump'' argues against the rapidrotation's being caused by the transport of internal angular momentum tothe surface. Capture of a planetary-mass companion as a red giantexpands in radius could explain the high rotational velocities. (4) Wealso find significant rotation in at least six of the roughly 15 (40%)red horizontal-branch stars in our survey. It is likely that theenhanced rotation seen among a significant fraction of both blue and redhorizontal-branch stars arose when these stars were luminous red giants.Rapid rotation alone therefore appears insufficient cause to populatethe blue side of the horizontal branch. While the largest projectedrotational velocities seen among field blue and red horizontal-branchstars are consistent with their different sizes, neither are consistentwith the large values we find for the largest red giants. This suggeststhat some form of angular momentum loss (and possibly mass loss) hasbeen at work. Also puzzling is the apparent absence of rotation seen infield RR Lyrae variables. Angular momentum transfer and conservation inevolved metal-poor field stars thus pose many interesting questions forthe evolution of low-mass stars. A grid of synthetic spectra and indices Fe5270, Fe5335, Mgb and Mg2 as a function of stellar parameters and [alpha/Fe]We have computed a grid of synthetic spectra in the wavelength rangelambda lambda 4600-5600 Å using revised model atmospheres, for arange of atmospheric parameters and values of [alpha -elements/Fe] = 0.0and +0.4. The Lick indices Fe5270, Fe5335, Mgb and Mg2 aremeasured on the grid spectra for FWHM = 2 to 8.3 Å. Relationsbetween the indices Fe5270, Fe5335 and Mg2 and the stellarparameters effective temperature Teff, log ; g, [Fe/H] and[alpha /Fe], valid in the range 4000 <= Teff <= 7000 K,are presented. These fitting functions are given for FWHM = 3.5 and 8.3Å. The indices were also measured for a list of 97 reference starswith well-known stellar parameters observed at ESO and OHP, and theseare compared to the computed indices. Finally, a comparison of theindices measured on the observed spectra and those derived from thefitting functions based on synthetic spectra is presented.Observations collected at the European Southern Observatory (ESO), LaSilla, Chile and at the Observatoire de Haute Provence (OHP), St-Michel,France.All Tables of Appendices A and B are only available in electronic format the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) orvia http://cdsweb.u-strasbg.fr/cgi-bin/qcat?/A+A/404/661 Oxygen line formation in late-F through early-K disk/halo stars. Infrared O I triplet and [O I] linesIn order to investigate the formation of O I 7771-5 and [O I] 6300/6363lines, extensive non-LTE calculations for neutral atomic oxygen werecarried out for wide ranges of model atmosphere parameters, which areapplicable to early-K through late-F halo/disk stars of variousevolutionary stages.The formation of the triplet O I lines was found to be well described bythe classical two-level-atom scattering model, and the non-LTEcorrection is practically determined by the parameters of theline-transition itself without any significant relevance to the detailsof the oxygen atomic model. This simplifies the problem in the sensethat the non-LTE abundance correction is essentially determined only bythe line-strength (Wlambda ), if the atmospheric parametersof Teff, log g, and xi are given, without any explicitdependence of the metallicity; thus allowing a useful analytical formulawith tabulated numerical coefficients. On the other hand, ourcalculations lead to the robust conclusion that LTE is totally valid forthe forbidden [O I] lines.An extensive reanalysis of published equivalent-width data of O I 7771-5and [O I] 6300/6363 taken from various literature resulted in theconclusion that, while a reasonable consistency of O I and [O I]abundances was observed for disk stars (-1 <~ [Fe/H] <~ 0), theexistence of a systematic abundance discrepancy was confirmed between OI and [O I] lines in conspicuously metal-poor halo stars (-3 <~[Fe/H] <~ -1) without being removed by our non-LTE corrections, i.e.,the former being larger by ~ 0.3 dex at -3 <~ [Fe/H] <~ -2.An inspection of the parameter-dependence of this discordance indicatesthat the extent of the discrepancy tends to be comparatively lessenedfor higher Teff/log g stars, suggesting the preference ofdwarf (or subgiant) stars for studying the oxygen abundances ofmetal-poor stars.Tables 2, 5, and 7 are only available in electronic form, at the CDS viaanonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/402/343 and Table\ref{tab3} is only available in electronic form athttp://www.edpsciences.org Abundances and Kinematics of Field Stars. II. Kinematics and Abundance RelationshipsAs an investigation of the origin of α-poor'' halo stars, weanalyze kinematic and abundance data for 73 intermediate-metallicitystars (-1>[Fe/H]>=-2) selected from Paper I of this series. We findevidence for a connection between the kinematics and the enhancement ofcertain element-to-iron ([X/Fe]) ratios in these stars. Statisticallysignificant correlations were found between [X/Fe] and galacticrest-frame velocities (vRF) for Na, Mg, Al, Si, Ca, and Ni,with marginally significant correlations existing for Ti and Y as well.We also find that the [X/Fe] ratios for these elements all correlatewith a similar level of significance with [Na/Fe]. Finally, we comparethe abundances of these halo stars against those of stars in nearbydwarf spheroidal (dSph) galaxies. We find significant differencesbetween the abundance ratios in the dSph stars and halo stars of similarmetallicity. From this result, it is unlikely that the halo stars in thesolar neighborhood, including even the α-poor'' stars, were oncemembers of disrupted dSph galaxies similar to those studied to date. Abundances of Cu and Zn in metal-poor stars: Clues for Galaxy evolutionWe present new observations of copper and zinc abundances in 90metal-poor stars, belonging to the metallicity range -3<[Fe/H]<-0.5. The present study is based on high resolutionspectroscopic measurements collected at the Haute Provence Observatoire(R= 42 000, S/N>100). The trend of Cu and Zn abundances as a functionof the metallicity [Fe/H] is discussed and compared to that of otherheavy elements beyond iron. We also estimate spatial velocities andgalactic orbital parameters for our target stars in order to disentanglethe population of disk stars from that of halo stars using kinematiccriteria. In the absence of a firm a priori knowledge of thenucleosynthesis mechanisms controlling Cu and Zn production, and of therelative stellar sites, we derive constraints on these last from thetrend of the observed ratios [Cu/Fe] and [Zn/Fe] throughout the historyof the Galaxy, as well as from a few well established properties ofbasic nucleosynthesis processes in stars. We thus confirm that theproduction of Cu and Zn requires a number of different sources (neutroncaptures in massive stars, s-processing in low and intermediate massstars, explosive nucleosynthesis in various supernova types). We alsoattempt a ranking of the relative roles played by different productionmechanisms, and verify these hints through a simple estimate of thegalactic enrichment in Cu and Zn. In agreement with suggestionspresented earlier, we find evidence that type Ia Supernovae must play arelevant role, especially for the production of Cu. Based on the spectracollected with the 1.93-m telescope of Haute Provence Observatory. Three-dimensional Spectral Classification of Low-Metallicity Stars Using Artificial Neural NetworksWe explore the application of artificial neural networks (ANNs) for theestimation of atmospheric parameters (Teff, logg, and [Fe/H])for Galactic F- and G-type stars. The ANNs are fed withmedium-resolution (Δλ~1-2 Å) non-flux-calibratedspectroscopic observations. From a sample of 279 stars with previoushigh-resolution determinations of metallicity and a set of (external)estimates of temperature and surface gravity, our ANNs are able topredict Teff with an accuracy ofσ(Teff)=135-150 K over the range4250<=Teff<=6500 K, logg with an accuracy ofσ(logg)=0.25-0.30 dex over the range 1.0<=logg<=5.0 dex, and[Fe/H] with an accuracy σ([Fe/H])=0.15-0.20 dex over the range-4.0<=[Fe/H]<=0.3. Such accuracies are competitive with theresults obtained by fine analysis of high-resolution spectra. It isnoteworthy that the ANNs are able to obtain these results withoutconsideration of photometric information for these stars. We have alsoexplored the impact of the signal-to-noise ratio (S/N) on the behaviorof ANNs and conclude that, when analyzed with ANNs trained on spectra ofcommensurate S/N, it is possible to extract physical parameter estimatesof similar accuracy with stellar spectra having S/N as low as 13. Takentogether, these results indicate that the ANN approach should be ofprimary importance for use in present and future large-scalespectroscopic surveys. Detection of Metal-poor Stars in the Direction of the North Galactic PoleA simple approach to detecting metal-poor stars is to measure amagnesium index, which depends on the Mg H band plus the three nearby Mgb lines and is derived through intermediate-band interference filters.An empirically established line of demarcation in the Mg index versusB-V diagram separates metal-poor stars from solar-abundance stars. Afurther separation between metal-poor dwarfs and giants depends on B-Vprimarily dwarfs for B-V<0.55, giants for B-V>0.7, with both dwarfsand giants falling in the transition region. For the metal-poor giantsthe distance from the demarcation line correlates well with [Fe/H],permitting estimates of stellar abundances. Stars in two regions on thesky in the vicinity of the north Galactic pole have been observed withsuch a set of filters. Eighteen stars (6% of the population of 299) inthe sample covering the V range 8.7 to 15.6 and 48 stars (31% of thepopulation of 163) in a deeper probe to V=19.9 found through thisprocess are suspected metal-poor stars according to their Mg indices.Twenty-three are specifically deemed giants, with<[Fe/H]><=-1.5. Analysis of neutron capture elements in metal-poor starsWe derived model atmosphere parameters (Teff, log g, [Fe/H],Vt) for 90 metal-deficient stars (-0.5<[Fe/H]<-3),using echelle spectra from the ELODIE library (Soubiran et al.\cite{soubet98}). These parameters were analyzed and compared withcurrent determinations by other authors. The study of the followingelements was carried out: Mg, Si, Ca, Sr, Y, Ba, La, Ce, Nd, and Eu. Therelative contributions of s- and r-processes were evaluated andinterpreted through theoretical computations of the chemical evolutionof the Galaxy. The chemical evolution models (Pagel &Tautvaišienė \cite{pagta95}; Timmes et al. \cite{timet95})depict quite well the behaviour of [Si/Fe], [Ca/Fe] with [Fe/H]. Thetrend of [Mg/Fe] compares more favourably with the computations of Pagel& Tautvaišienė (\cite{pagta95}) than those of Timmes etal. (\cite{timet95}). The runs of n-capture elements vs. metallicity aredescribed well both by the model of Pagel & Tautvaišienė(\cite{pagta95}, \cite{pagta97}) and by the model of Travaglio et al.(\cite{travet99}) at [Fe/H]>-1.5, when the matter of the Galaxy issufficiently homogeneous. The analysis of n-capture element abundancesconfirms the jump in [Ba/Fe] at [Fe/H]=-2.5. Some stars from our sampleat [Fe/H]<-2.0 show a large scatter of Sr, Ba, Y, Ce. This scatter isnot caused by the errors in the measurements, and may reflect theinhomogeneous nature of the prestellar medium at early stages ofgalactic evolution. The matching of [Ba/Fe], [Eu/Fe] vs. [Fe/H] with theinhomogeneous model by Travaglio et al. (\cite{travet01a}) suggests thatat [Fe/H]<-2.5, the essential contribution to the n-rich elementabundances derives from the r-process. The main sources of theseprocesses may be low mass SN II. The larger dispersion of s-processelement abundances with respect to alpha -rich elements may arise bothfrom the birth of metal-poor stars in globular clusters with followingdifferent evolutionary paths and (or) from differences in s-elementenrichment in Galaxy populations. Based on spectra collected at theObservatoire de Haute-Provence (OHP), France On the stellar content of the open clusters Melotte 105, Hogg 15, Pismis 21 and Ruprecht 140CCD observations in the B, V and I passbands have been used to generatecolour-magnitude diagrams reaching down to V ~ 19 mag for two slightlycharacterized (Melotte 105 and Hogg 15) and two almost unstudied (Pismis21 and Ruprecht 140) open clusters. The sample consists of about 1300stars observed in fields of about 4arcmin x4arcmin . Our analysis showsthat neither Pismis 21 nor Ruprecht 140 are genuine open clusters sinceno clear main sequences or other meaningful features can be seen intheir colour-magnitude diagrams. Melotte 105 and Hogg 15 are openclusters affected by E(B-V) = 0.42 +/- 0.03 and 0.95 +/- 0.05,respectively. Their distances to the Sun have been estimated as 2.2 +/-0.3 and 2.6 +/- 0.08 kpc, respectively, while the corresponding agesestimated from empirical isochrones fitted to the Main Sequence clustermembers are ~ 350 Myr and 300 Myr, respectively. The present data arenot consistent with the membership of the WN6 star HDE 311884 to Hogg15. Tables 2 to 5 are only available in electronic form at the CDS viaanonymous ftp to cdsarc.u-strasbg.fr (130.793.128.5) or viahttp://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/370/931 Neutron-Capture Elements in the Early Galaxy: Insights from a Large Sample of Metal-poor GiantsNew abundances for neutron-capture (n-capture) elements in a largesample of metal-poor giants from the Bond survey are presented. Thespectra were acquired with the KPNO 4 m echelle and coudé feedspectrographs, and have been analyzed using LTE fine-analysis techniqueswith both line analysis and spectral synthesis. Abundances of eightn-capture elements (Sr, Y, Zr, Ba, La, Nd, Eu, and Dy) in 43 stars havebeen derived from blue (λλ4070-4710, R~20,000, S/Nratio~100-200) echelle spectra and red (λλ6100-6180,R~22,000, S/N ratio~100-200) coudé spectra, and the abundance ofBa only has been derived from the red spectra for an additional 27stars. Overall, the abundances show clear evidence for a largestar-to-star dispersion in the heavy element-to-iron ratios. Thiscondition must have arisen from individual nucleosynthetic events inrapidly evolving halo progenitors that injected newly manufacturedn-capture elements into an inhomogeneous early Galactic halointerstellar medium. The new data also confirm that at metallicities[Fe/H]<~-2.4, the abundance pattern of the heavy (Z>=56) n-captureelements in most giants is well-matched to a scaled solar systemr-process nucleosynthesis pattern. The onset of the main r-process canbe seen at [Fe/H]~-2.9 this onset is consistent with the suggestion thatlow mass Type II supernovae are responsible for the r-process.Contributions from the s-process can first be seen in some stars withmetallicities as low as [Fe/H]~-2.75 and are present in most stars withmetallicities [Fe/H]>-2.3. The appearance of s-process contributionsas metallicity increases presumably reflects the longer stellarevolutionary timescale of the (low-mass) s-process nucleosynthesissites. The lighter n-capture elements (Sr-Y-Zr) are enhanced relative tothe heavier r-process element abundances. Their production cannot beattributed solely to any combination of the solar system r- and mains-processes, but requires a mixture of material from the r-process andfrom an additional n-capture process that can operate at early Galactictime. This additional process could be the weak s-process in massive(~25 Msolar) stars, or perhaps a second r-process site, i.e.,different from the site that produces the heavier (Z>=56) n-captureelements. Abundances and Kinematics of Field Halo and Disk Stars. I. Observational Data and Abundance AnalysisWe describe observations and abundance analysis of a high-resolution,high signal-to-noise ratio survey of 168 stars, most of which aremetal-poor dwarfs. We follow a self-consistent LTE analysis technique todetermine the stellar parameters and abundances, and we estimate theeffects of random and systematic uncertainties on the resultingabundances. Element-to-iron ratios are derived for key α-, odd-Z,Fe-peak, and r- and s-process elements. Effects of non-LTE on theanalysis of Fe I lines are shown to be very small on average.Spectroscopically determined surface gravities are derived that arequite close to those obtained from Hipparcos parallaxes. Galactic [O/Fe] and [C/Fe] Ratios: The Influence of New Stellar ParametersWe consider the effects of recent NLTE gravities and Fe abundances onstellar [O/Fe] and [C/Fe] ratios. The NLTE parameters greatly reduce oreliminate the well-known discrepancy between CH- and C I-based Cabundances in metal-poor stars and previously seen trends ofatomic-based [C/Fe] and [O/Fe] with Teff. With the NLTEparameters, the metal-poor molecular-based [C/Fe] ratio maintains itsincrease with declining [Fe/H] this may also be demonstrated by therevised atomic-based ratios. [O/Fe] values derived from OH and O Ifeatures are considerably reduced and typically show improved agreementbut are 0.1-0.2 dex larger than those exhibited by the Lick-Texassyndicate's recent [O I]-based giant determinations. The revised [O/Fe]ratios still show an increase down to at least [Fe/H]~-2 we suggest thatrecent field giant data show an increase with similar slope. Evenadopting uniform NLTE parameters, study-to-study abundance differencescan be significant; moreover, different NLTE studies yield differinggravities and Fe abundances even after taking Teffdifferences into account. Comparison of metal-poor giant gravities andcluster abundances with isochrones, trigonometric gravities, andnear-turnoff cluster abundances yields conflicting indications aboutwhether the evolved gravities might be underestimated as suggested formetal-poor dwarfs. Regardless, we argue that even extreme gravityrevisions do not affect the [O/Fe]-[Fe/H] relation derived from theextant results. Combining what we believe the most reliable giant anddwarf data considered here, we find[O/Fe]=-0.184(+/-0.022)×[Fe/H]+0.019 with an rms scatter of only0.13 dex; there is no indication of a break or slope change atintermediate [Fe/H]. The gentle slope is in very reasonable agreementwith some chemical evolution models employing yields with small mass andmetallicity dependences. Finally, two notes are made concerning Naabundance-spatial position and element-to-element correlations in M13giants. Kinematics of Metal-poor Stars in the Galaxy. II. Proper Motions for a Large Nonkinematically Selected SampleWe present a revised catalog of 2106 Galactic stars, selected withoutkinematic bias and with available radial velocities, distance estimates,and metal abundances in the range -4.0<=[Fe/H]<=0.0. This updateof the 1995 Beers & Sommer-Larsen catalog includes newly derivedhomogeneous photometric distance estimates, revised radial velocitiesfor a number of stars with recently obtained high-resolution spectra,and refined metallicities for stars originally identified in the HKobjective-prism survey (which account for nearly half of the catalog)based on a recent recalibration. A subset of 1258 stars in this cataloghave available proper motions based on measurements obtained with theHipparcos astrometry satellite or taken from the updated AstrographicCatalogue (second epoch positions from either the Hubble Space TelescopeGuide Star Catalog or the Tycho Catalogue), the Yale/San Juan SouthernProper Motion Catalog 2.0, and the Lick Northern Proper Motion Catalog.Our present catalog includes 388 RR Lyrae variables (182 of which arenewly added), 38 variables of other types, and 1680 nonvariables, withdistances in the range 0.1 to 40 kpc. Oxygen abundance in halo stars from O i tripletOxygen abundance for 14 halo stars through the O I 7774 Ätriplethave been derived from high resolution spectra (R = 25,000; S/N >100) obtained with echelle-spectrometer of 6-m telescope of SpecialAstrophysical Observatory of the Russian Academy of Sciences. Theeffective temperature, metallicity and other parameters have beenexamined. For example, the effective temperature was found from H_alphaline wings and photometric indices. The abundance analysis was carriedout using both LTE and non-LTE conceptions. For this aim, we havespecified the oxygen atomic model. The average [O/Fe] value appeared tobe 0.61 +/- 0.21 from the non-LTE determination. A trend of oxygenabundance increasing along with the iron abundance decreasing was found.The relation between [O/Fe] and [Fe/H] is linear:[O/Fe]=-0.370x[Fe/H]+0.047. In addition to the sample of our programstars, we also involved in the analysis, 24 targets from Cavallo et al.(\cite{Cav}). For their original results we have determined thenecessary non-LTE corrections. Our data are compared with the results ofother works (Tomkin et al., \cite{Tom}; King & Boesgaard,\cite{King2}; Boesgaard et al. \cite{BK2}). Radial Velocities of Population II Stars. II.A program for radial velocity measurements of Population II stars wasstarted in 1988 and was carried out during six observing runs. Theprogram includes metal-deficient stars, components of Population IIvisual binaries or common proper motion stars, suspected radial velocityvariables and the Population II stars from the Hipparcos program. Themeasurements were made with the 1 meter reflector at the MaidanakObservatory in Uzbekistan. The average error of a single measurement isabout 0.6 km/s, but for stars at 13 mag or for extremely metal-deficientstars the error is about 2.5 km/s. The catalog contains 621 measurementsfor 164 stars. Estimation of Stellar Metal Abundance. II. A Recalibration of the Ca II K Technique, and the Autocorrelation Function MethodWe have recalibrated a method for the estimation of stellar metalabundance, parameterized as [Fe/H], based on medium-resolution (1-2Å) optical spectra (the majority of which cover the wavelengthrange 3700-4500 Å). The equivalent width of the Ca II K line (3933Å) as a function of [Fe/H] and broadband B-V color, as predictedfrom spectrum synthesis and model atmosphere calculations, is comparedwith observations of 551 stars with high-resolution abundances availablefrom the literature (a sevenfold increase in the number of calibrationstars that were previously available). A second method, based on theFourier autocorrelation function technique first described by Ratnatunga& Freeman, is used to provide an independent estimate of [Fe/H], ascalibrated by comparison with 405 standard-star abundances.Metallicities based on a combination of the two techniques for dwarfsand giants in the color range 0.30<=(B-V)_0<=1.2 exhibit anexternal 1 sigma scatter of approximately 0.10-0.20 dex over theabundance range -4.0<=[Fe/H]<=0.5. Particular attention has beengiven to the determination of abundance estimates at the metal-rich endof the calibration, where our previous attempt suffered from aconsiderable zero-point offset. Radial velocities, accurate toapproximately 10 km s^-1, are reported for all 551 calibration stars.
Submit a new article