Physical characterization of Galactic O-type stars targeted by the IACOB and OWN surveys
HHighlights on Spanish Astrophysics IX, Proceedings of the XII Scientific Meeting of the SpanishAstronomical Society held on July 18 – 22, 2016, in Bilbao, Spain. F. Figueras, A. Sánchez-Lavega, S.Pérez-Hoyos, A. Alonso Herrero, S. Arribas, C. Hernández Monteagudo (eds.)
Physical characterization of Galactic O-type starstargeted by the IACOB and OWN surveys
Gonzalo Holgado , , Sergio Sim´on-D´ıaz , , and Rodolfo Barb´a Instituto de Astrof´ısica de Canarias, E-38200 La Laguna, Tenerife, Spain. Departamento de Astrof´ısica, Univ. de La Laguna, E-38205 La Laguna, Tenerife, Spain. Departamento de F´ısica y Astronom´ıa, Univ. de la Serena, Av. Juan Cisternas 1200Norte, La Serena, Chile
Abstract
We present first results from the quantitative spectroscopic analysis of ∼
270 Galactic O-typestars targeted by the IACOB and OWN surveys (implying the largest sample of stars of thistype analyzed homogeneously). We also evaluate what is the present situation regardingavailable information about distances, as provided by the Hipparcos and Gaia missions.
We are immersed in the era of the large spectroscopic surveys of massive O- and B-type stars.An era which is smoothly overlapping with the time in which the Gaia mission is expectedto provide fresh momentum to the field of stellar astrophysics.Based on the exhaustive Galactic O-star Catalogue ([1]), the IACOB ([2, 3, 4]) andOWN ([5, 6]) projects have independently devoted an enormous observational effort in thelast decade to compile two large multi-epoch high-resolution spectroscopic databases of Galac-tic O-type stars, including Northern and Southern targets, respectively. A few years ago, wedecided to joint efforts and start the exploitation of this unique spectroscopic dataset re-garding the physical characterization of the stellar properties of the complete sample. Tosuccessfully reach our objective we are benefiting from a battery of self-made automatizedtools optimized for the quantitative spectroscopic analysis of large samples of O-type stars,and plan to use the upcoming information about distances provided by the ESA-Gaia mission([7]). In this proceeding we introduce the main sample which is being analyzed and presentfirst results of our on-going work. a r X i v : . [ a s t r o - ph . S R ] N ov Physical characterization of Galactic O-type stars from IACOB and OWN
Table 1: Summary of the spectroscopic observations used in this workSurvey Telescope Instrument Resolution Range [˚A]
We used the Galactic O-star Catalogue (GOSC, P.I., J. Ma´ız Apell´aniz) as initial reference tobuild our sample of stars to be analyzed spectroscopically. The GOSC has presently becomethe most complete and updated compilation of accurate information related to this type ofobjects. Its current public version comprises 601 Galactic stars with spectral types in therange O2 – 09.7 (all luminosity classes) from both – Northern and Southern – hemispheres(see Ma´ız Apell´aniz, these proceedings).We searched in the two modern multi-epoch, high resolution spectroscopic databasesIACOB (P.I., S. Sim´on-D´ıaz), and OWN (P.Is., R. Barb´a & R. Gamen), and found a total of2452 spectra for 340 of the stars listed in GOSC. In particular, we concentrated on the spectracollected by these surveys with three different instruments: FIES, HERMES, and FEROS.The similarity between these spectroscopic observations – in terms of resolving power andS/N – allows to minimize observational effects on the outcome from our spectroscopic study.A global overview of the number of stars and spectra available from each survey is presentedin Table 1.As a natural consequence of the observing philosophy of the IACOB and OWN sur-veys, most of the stars count on more than 2 spectra obtained at different epochs. Onlythe spectrum with the best S/N was considered for the quantitative spectroscopic analysisleading to the stellar parameters, but we used the complete spectroscopic dataset to obtaininformation about spectroscopic binarity or variabilityThe initial sample of stars with high resolution spectra represents ∼
56% of the GOSCsample. This is due to the more limited range in magnitude characterizing the stars targetedby IACOB and OWN (see Fig. 1). We note, however, that if we just concentrate in starsbrighter than B ap =9 (where IACOB and OWN have been mainly concentrated to date) thehigh resolution sample is complete up to ∼ Part of these observations correspond to the IACOBsweg program [4] led by I.Negueruela In this case, also including available spectra from the CAFE-BEANS survey [8] An extension of the IACOB and OWN observations with the objective of being complete up to V=11 isalready on-going. olgado et al. ap GOSCIACOB+OWN
Figure 1: B ap magnitude histogram the comparing the GOSC and IACOB+OWN samples. We performed the quantitative spectroscopic analysis of our final list of 266 Galactic O-typestars using several tools developed in the framework of the IACOB project. First, iacob-broad [9] was used to obtain the projected rotational velocity ( v sin i ) and the amount ofnon-rotational broadening (aka. macroturbulence, v mac ) affecting the line-profiles of each star(i.e. the line-broadening parameters ). Then, iacob-gbat [10], a semi-automatized tool forthe quantitative spectroscopic analysis of O-stars based on an extensive grid of fastwind [11] models and an optimized χ strategy, was used to obtain the so-called spectroscopicparameters ; namely, the effective temperature ( T eff ), gravity (log g ), wind-strength parameter( Q ), helium abundance ( Y He ), microturbulence ( ξ t ), and the exponent of the wind velocity-law ( β ). We refer the reader to [9], [10], and [12] for a thorough description of the tools andsome of the basic ideas about the analysis strategy and the interpretation of results.The information extracted from the spectra will be complemented with that for theother fundamental parameters – such as the radius ( R ), the luminosity ( L ), or the spectro-scopic mass ( M sp ) – once reliable and accurate information about distances will be madeavailable by Gaia ( see also Sect. 5). We present in this section an overview of the first results obtained from the quantitativespectroscopic analysis of the considered sample of Galactic stars. As indicated above, thisrefers to 266 O-type stars not detected as double line spectroscopic binaries from inspection
Physical characterization of Galactic O-type stars from IACOB and OWN v sin i [kms ] 01020304050 I (66)II (25)III (52)IV (39)V (84) i [kms ]01020304050 Likely single (227)SB1 (39)
Figure 2: Distribution of projected rotational velocities for a sample of 266 O-type stars,separated by ( left) luminosity class, ( right) detected binarity status. We note that results for ∼
75 SB2 stars have been removed from this graph.of the available FIES, HERMES and FEROS from the IACOB and OWN surveys. Thisimplies the larger sample of Galactic O-type stars spectroscopically analyzed to date in ahomogeneous way.
Projected rotational velocities
In Fig. 2 we show the distribution of the projected rotational velocities in the analyzed sampleof O-type stars, separated by luminosity class (left) and binary status (right). A thoroughstudy of these distributions will allow us to step forward in our knowledge about how thisimportant stellar parameter is initially distributed and is modified as the star evolves and/orinteract with other components in a binary system. Also, by comparing this result with thatobtained by, e.g., [13], we will be able to investigate the effect of metallicity on the evolutionof spins rates in massive stars.
Stellar properties in the spectroscopic HR diagram
In Fig. 3 we present how luminosity class, spectral type, as well as four of the stellar pa-rameters determined spectroscopically ( v sin i , Y ( He ), log Q , and ξ t ) are distributed in thespectroscopic HR diagram [14]. This figure shows the good coverage of the classic O-typedomain we have reached with the compiled and analyzed sample of stars. We are confidentthat our study (once finalized by incorporating information about radii, luminosities, spec-troscopic masses and abundances) will provide an unprecedented empirical overview of themain physical properties of Galactic massive O-type stars to be used as definitive anchorpoint for our theories of stellar atmospheres, winds, interiors and evolution of massive stars. olgado et al. a) IIIIIIIVV 15202532406085120 b) SpTO6O7.5O8.5O9.7 e) T eff [kK] log Q [dex]< - 13.5-13.5:-13.0-13.0:-12.5-12.5:-12.0>-12.0 f) ξ t [kms − ]< 55:1010:15> 15 l o g L / L fl c) v sini [kms − ]< 4040:80> 80 15202532406085120 d) Y (He)< 88:1212:15> 15
Figure 3: Distribution in the spectroscopic HR diagram of several stellar properties deter-mined spectroscopically in our sample of 266 likely single and/or SB1 O-type stars. Openpoints are limits. Non rotating evolutionary tracks (and ZAMS) from the Geneva group ([17]and [18]) overplotted for reference.
Physical characterization of Galactic O-type stars from IACOB and OWN / C u m u l a t i v e d i s t r i b u t i o n f un c t i o n Figure 4: Normalized cumulative distribution function of the relative error in TYCHO andGAIA-TGAS parallaxes for our sample of 340 Galactic O-type stars. Negative parallaxeshave been eliminated.
As indicated above, our final objective is the complete characterization of the whole sampleof Galactic O-type stars targeted by IACOB and OWN, also including information aboutabundances, as well as the remaining fundamental parameters: L , R , and M sp . For the later,accurate distances are needed.Where are we in terms of distances? To answer this question, we searched for availableinformation about parallaxes for our initial sample of 340 stars. We collected 152 and 216parallaxes with positive values in the TYCHO catalog [15] and the recently released GAIA-TGAS DR1 [16], respectively. The compiled information is summarized in Fig. 4 as twonormalized cumulative distribution functions of the relative error resulting from each dataset.Overall, there is a clear improvement of the situation from TYCHO to TGAS, with a highernumber of parallaxes with positive values and a decrease in the mean relative error; however,we have not reach yet a situation good enough for the purposes of our work since still only ∼
40% of the sample present a formal relative error (as provided in TGAS) below 50%.We will hence have to wait for GAIA DR2, when an important improvement in theparallax determination and the associated uncertainties is expected, to be able to computereliable values for the physical parameters of our sample.
References [1] Ma´ız Apell´aniz, J., Sota, A., Morrell, N. I., et al. 2013, Massive Stars: From alpha to Omega, 198[2] Sim´on-D´ıaz, S., Garcia, M., Herrero, A., Ma´ız Apell´aniz, J., & Negueruela, I. 2011a, SCA: A RIA olgado et al.7