EDP Sciences
Free Access
Issue
A&A
Volume 581, September 2015
Article Number A68
Number of page(s) 20
Section Catalogs and data
DOI https://doi.org/10.1051/0004-6361/201424155
Published online 04 September 2015

© ESO, 2015

1. Introduction

The Convection, Rotation, and planetary Transits (CoRoT) space mission (Baglin et al. 2007) collected a total of 161 303 point-source photometric data over a period of six years for stars exhibiting different luminosity classes and spectral types. This space mission had two main goals: 1) the detection of extra-solar planets using the transit procedure; and 2) precise stellar seismology. In addition to these two big challenges, other programs related to the CoRoT mission are helping further our understanding of a variety of important astrophysical phenomena, such as stellar activity, pulsation, and multiplicity. In this context, CoRoT provides a unique opportunity for the study of stellar rotation, which is recognized as a fundamental quantity that controls the evolution of stars, however, it is only today that the models and observations in hand to begin to address it. In this sense, CoRoT offers the necessary tools for the photometric measurements of rotation periods for a statistically robust sample of stars at different evolutionary stages and belonging to different stellar populations.

However, in spite of the large amount of high-quality photometric data obtained with CoRoT, ground-based observations are also needed to accomplish the main scientific goals of the mission. For this reason, several observational programs have increased the photometric database related to the CoRoT mission (e.g., Aigrain et al. 2009; Deleuil et al. 2009). These data enabled the establishment of the best fields and observation setups to observe with the satellite, also establishing the spectral types and luminosity classes for the stars in the CoRoT fields. Nevertheless, important uncertainties are still present in these classifications because of the variable reddening levels affecting the CoRoT fields, in addition to the still unknown chemical abundances and distances to the targets.

Spectroscopic observations are mandatory for a solid treatment of the different CoRoT scientific goals. In this sense, two large spectroscopic surveys of CoRoT targets have been carried out to date, both using multifiber observations. The first survey (Gazzano et al. 2010, G10 hereafter) combined multifiber observations with an automated procedure for the determinations of different stellar parameters, whereas the second was dedicated essentially to spectral classification (Sebastian et al. 2012).

In the context of the physical characterization of CoRoT targets, we carried out a large spectroscopic survey focused on the brightest F-, G-, and K-type stars in the CoRoT exoplanet fields LRc01 and LRa01, using the multifiber spectrographs UVES/VLT and Hydra/Blanco, with high and medium spectral resolution, respectively. Using these observations, we applied a homogeneous procedure for the determination of different stellar parameters, including effective temperature (Teff), surface gravity (log  (g)), overall metallicity ([Fe/H]), radial velocity (vrad), projected rotational velocity (vsin (i)), and microturbulence (vmic). The main goal of this paper is to present the corresponding catalog. We also present the mean values for stellar parameters of the two stellar populations in the CoRoT anticenter/center direction. The paper is structured as follows: in Sect. 2 we describe the observations and data reduction. Section 3 describes how we derived the stellar parameters for the stars in our sample, and Sect. 4 contains our main results. Finally, we draw our conclusions in Sect. 5.

2. Observations

The present stellar sample is composed of 138 stars of spectral types F, G, and K, with visual magnitudes V between 10 to 14, located in two exoplanet fields observed by CoRoT, namely the Galactic center (LRc01: Long Run Center 01) and the Galactic anticenter (LRa01: Long Run Anticenter 01) fields. We selected the sample using as criteria the visual magnitude V, the spectral type, and the luminosity classes defined by Deleuil et al. (2009) for CoRoT targets. We selected stars belonging to luminosity classes II, III, IV, and V considering the range in V and spectral type defined above. Our sample is comprised of the brightest stars in both CoRoT fields, and is thus not fully representative of the magnitude and color distribution of CoRoT stars.

To obtain a physical characterization for these stars, a series of spectroscopic observations were carried out using two spectrographs. A sample of 53 stars was observed using the high-resolution UVES spectrograph (hereafter UVES stars) mounted on the Kuyen/VLT 8.2 m telescope, located in Cerro Paranal, Chile, in the course of different observing runs in 2006. The UVES standard setup DICH-2 (390–760 nm) with a 0.9 arcsec slit was used, allowing us to obtain high-resolution (R ≈ 47 000) and high signal-to-noise (S/N> 100) spectra. The main characteristics of the targets and the observation dates are given in Table 1.

A complementary sample of 85 stars was observed using the Hydra multifiber echelle spectrograph (henceforth Hydra stars), mounted on the Blanco 4m telescope at the Cerro Tololo Interamerican Observatory, located in Cerro Tololo, Chile. The filters E5187 (509–525 nm) and E6757 (656–681 nm) with a 200 micron slit were used, allowing us to collect spectra with medium resolution (R ≈ 17 000 and 15 000, respectively) and signal-to-noise ratio 70 <S/N< 200. The E5187 filter was chosen because it covers a spectral region with five Fe II lines, whereas the E6757 was chosen because several Fe I lines and a Li doublet (at 671 nm) are located in spectral window. Also, a few stars, with accurate previous measurements of vsin (i) and with FGK spectral types, were also observed with the same Hydra setups (Melo et al. 2001) to construct a calibration for the determination of vsin (i) for the Hydra stars in our sample. Figure 1 shows examples of these observations using both instruments.

Table 2 shows the setup used in Hydra observations. For the targets, we also compiled luminosity classes, V-band magnitudes, and color indices from the CoRoT database1, as well as JHKs magnitudes from the Two-Micron All-Sky Survey (2MASS) database (Skrutskie et al. 1996). The main characteristics of the targets and the observation dates, and corresponding luminosity classes (from the CoRoT and 2MASS databases) are given in Tables 3 and 4.

thumbnail Fig. 1

Some spectra of the stars contained in the sample. Spectra collected using the UVES spectrograph are show in the upper panel, whereas the lower panel shows spectra collected using the Hydra spectrograph.

Open with DEXTER

Table 1

CoRoT IDs, ephemerides, details of observations, photometry, and estimated luminosity classes (LC) for the UVES stars.

Table 2

Setup used for Hydra observations.

The reduction of UVES stars data was done using the standard UVES data reduction pipeline (Ballester et al. 2000). Hydra data were reduced using the dohydra task in IRAF2. Both reductions follow the usual reduction steps (bias, flat-field, and background corrections, fiber order definition, wavelength calibration of the spectra with Th-Ar lamp spectra, and extraction of the spectra). Then we use IRAF to normalize the spectra to a pseudocontinuum level and to bring the reduced spectra to the rest frame. Cosmic rays were extracted using the procedures described in van Dokkum (2001).

3. Stellar properties

3.1. Radial velocities vrad

We obtained radial velocities vrad with the fxcor task (Tonry & Davis 1979) in IRAF. Because the stars of the UVES sample present Teff close to the solar value, we cross-correlated the UVES spectra with a spectrum of the Sun (Hinkle et al. 2000). We then converted the shifts into radial velocities of the stars, and we applied a barycentric correction. On the other hand, because the Hydra stars present a greater spread in temperatures and luminosity classes, the spectra were cross-correlated with synthetic spectra of the Sun and an RGB star (Teff = 4000 K, log  (g) = 1.0 dex and [Fe/H] = 0.0 dex) to compare differences in the determinations of radial velocities. We computed the spectra with the Turbospectrum code (Alvarez & Plez 1998) and MARCS atmosphere models with solar abundances (Gustafsson et al. 2008). In Fig. 2 small differences can be found between the radial velocities derived using the synthetic spectra (averaging about −0.27 ± 0.37km s-1). We opted to use the values found using the synthetic solar spectrum, and applied a barycentric correction. The typical errors in radial velocities for Hydra stars are lower than 0.5km s-1. On the other hand, we also computed a synthetic solar spectrum for the UVES spectral resolution, which we used to obtain vrad for the UVES sample, to find systematic errors in the Hydra vrad measured using synthetic spectra. We found that our measurements of vrad using synthetic spectra present a systematic difference of about −0.75 ± 1.66km s-1 with those derived with observed solar spectrum, and is thus not significant.

Table 3

Main characteristics for the Hydra stars.

Table 4

Main characteristics and rotation velocities vsin (i) of calibrator stars.

thumbnail Fig. 2

Comparison between the vrad measurements for the Hydra sample obtained using different synthetic spectra. The represents the difference between the vrad obtained when the Hydra spectra are cross-correlated with synthetic spectra for a RGB star and the Sun, respectively. The red line represents the mean value of and the standard deviation is presented using black dashed lines. Also, the typical error in vrad measurements for Hydra stars is shown using an error bar.

Open with DEXTER

3.2. Rotation velocities v sin (i)

The vsin (i) measurements of our targets were computed using two procedures. For the case of UVES stars, the vsin (i) values were determined using the same procedure as in Canto Martins et al. (2011). Following these authors, the resulting spectra (taking into account the instrumental profile of UVES) are convolved with rotational profiles to adjust the broadening observed in the iron lines (profile fitting) located between 6700 and 6715 Å.

The vsin (i) values for the Hydra stars were computed (using the fxcor task) with a cross-correlation function (CCF) especially calibrated for the Hydra spectrograph. We then followed the same procedure decribed before for the UVES stars. As is described in Recio-Blanco et al. (2002) and Lucatello & Gratton (2003), the relation between vsin (i) and the corrected width σobs0 of the CCF is (1)where A and σ0 are the so-called coupling constant and the nonrotational contribution to the CCF width, respectively. As mentioned in Melo et al. (2001), σ0 depends on different broadening mechanisms (magnetic field, instrumental profile, thermal broadening, etc.), which is related to object star and the template used, but does not depend on rotation.

Since the setup E5187 presents the highest resolution in our Hydra observations, it was chosen to determine the vsin (i) values. We used FGK stars with reliable vsin (i) determinations as templates and calibrators, which were observed during our observing runs. The vsin (i) values and photometry for these stars are compiled in Table 4.

The fxcor task allowed us to obtain the uncorrected width of the CCF (σobs), which has a contribution from the template used (σt) in deriving the CCF. The σt can be determined with an autocorrelation for each template, as is described in Eq. (4) of Lucatello & Gratton (2003). For each star used as template we obtained several spectra, which allowed us to avoid the autocorrelation of the same spectrum. The σobs and σt are related with the corrected width σobs0 through the following equation: (2)The mean values of σobs and σt for each template are listed in Table 5, whereas the mean values of σobs for each calibrator star are listed in Table 6. Finally, using a linear fit in the plane [(σobs0)2,(vsin(i))2] the following relation between vsin (i) and σobs0 was obtained: The errors in these coefficients are associated with the errors in the slopes of the linear fit. In Fig. 3 we show the final calibration, which presents a good agreement with the reference values of vsin (i) for calibrator stars.

Table 5

Observed broadening σobs and σt for template stars.

thumbnail Fig. 3

CCF-vsin (i) calibration for the Hydra spectrograph. The dashed line represents the function found and black points represent the values for the calibrator stars, respectively.

Open with DEXTER

Table 6

Observed broadening σobs and σobs0 for calibrator stars.

These vsin (i) values were used as reference to obtain new measurements of vsin (i) values using the setup E6757 and the same method described for UVES stars. Note that the setup E6757 contains the spectral region between 6700 and 6715 Å. Small differences were found between vsin (i) values derived using both methods. Since we use the profile fitting to derive lithium abundances (as is described in Sect. 4.4), the final values of vsin (i) for the Hydra stars are those derived from the profile fitting.

3.3. Effective temperatures Teff, surface gravities log (g), iron abundance [Fe/H], and microturbulence velocity vmic

Because our stellar sample is comprised of stars belonging to the field, it is important to have an estimation of their stellar parameters, which were used to avoid mistakes in the determination of the final parameters. In this sense, as a first step in the derivation of atmospheric parameters, we used the 2MASS JHK near-infrared photometry and the CoRoT database mean values of V and ⟨ (BV) ⟩ to obtain a first estimation of the effective temperature for our sample. Specifically, we used the mean CoRoT color index (BV) and the calibrations of Flower (1996) corrected by Torres (2010) to calculate the photometric effective temperature Teff(BV). In the same way, we calculated the photometric effective temperature Teff(JK) using the 2MASS color index (JK), the CoRoT luminosity classes, and the calibrations of Alonso et al. (1996, 1999). We derived these temperatures without reddening corrections (see Sect. 4.3 for detailed discussion). Also, we found errors at levels of 0.04 mag in the (JK) for the stars in our sample, which implies errors at levels of 130 K in Teff(JK). No errors are informed in the CoRoT database for (BV).

We used the average between both photometric temperatures as our initial estimation for the spectroscopic temperature. At the same time, we estimated the initial values of surface gravities log  (g) using the CoRoT luminosity classes and interpolations in tables of infrared synthetic colors computed with ATLAS9 by R. Kurucz3. These initial estimation can present important errors, produced by reddening and bad identification of CoRoT luminosity classes (see Fig. 8 in G10). In fact we found differences of about >500 K between photometric and spectroscopic temperatures, which implies high rates of extinction in the CoRoT fields (see Sect. 4.3). For this reason we stress that these initial values were used only as a starting point to obtain the final parameters and, in any case, they constrained the searching of spectroscopic temperatures.

We determined final values of the atmospheric parameters and their respective errors using the Turbospectrum code and MARCS atmosphere models with solar abundances. Solar abundances were taken from Asplund et al. (2005), and the collisional damping treatment was performed based on the work of Barklem and co-workers (Barklem et al. 2000a,b; Barklem & Piskunov 2003; Barklem & Aspelund-Johansson 2005). To compute synthetic spectra with the Turbospectrum code, we took atomic (see below) and molecular line lists into account, including TiO (Plez 1998), VO (Alvarez & Plez 1998), and CN and CH (Hill et al. 2002). The Turbospectrum code uses equivalent widths (EW) to compute abundances A(Fe) corresponding to the different Fe lines4. The list of Fe lines used was compiled and corrected by Canto Martins et al. (2011, see their Table 6). This list is composed of 91 and 14 lines of Fe I and II, respectively. There are differences in the number of iron lines used to characterize the stars because our sample was observed using two instruments and different setups. In particular, for UVES stars, we used all lines compiled by Canto Martins et al. (2011), whereas for the Hydra stars 33 of the lines (28 Fe I lines and 5 Fe II lines) in the intervals 5090 − 5250 Å and 6560 − 6810 Å were used. We measured EW values with the DAOSPEC code (Stetson & Pancino 2008). Using excitation equilibrium for the Fe I abundances, Fe I/Fe II ionization equilibrium, and the equilibrium of the A(Fe) values and their respective EW values, we can derive effective temperatures Teff, surface gravities log  (g), and microturbulence velocities vmic, respectively. Starting from the photometric parameters, we ran the Turbospectrum code iteratively using MARCS atmosphere models with different parameters to find the three equilibria, thus defining the final parameters. Figure 4 presents an example of physical and chemical parameter determinations using the equilibria (slopes equal to zero in the planes shown in this figure) for both a UVES star and a Hydra star.

thumbnail Fig. 4

Upper panels: ionization equilibrium for the targets 100932329 (UVES star) and 101565091 (Hydra star). Lower panels: equilibrium of the A(Fe)s and their EWs. Open and filled circles represent the abundances of Fe I (A(FeI)) and Fe II (A(FeII)), respectively. Red and black dashed lines represent the mean value of A(Fe) and standard deviation, respectively.

Open with DEXTER

The errors in Teff and vmic are obtained from the errors in the slopes that define the equilibria described above. We change one of these parameters, keeping the others fixed, and compute a new atmospheric model. We found the error when the slope of the new fit became equal to its respective slope error. The error in [Fe/H] is equal to slope error in excitation equilibrium for the Fe I abundance. Finally, the error in log  (g) is found when the difference between A(FeI) and A(FeII) is equal to the square root of the sum of the squares of the errors in A(FeI) and A(FeII).

3.4. The Li abundance determinations

The Li abundances A(Li)5 for UVES and Hydra stars were calculated by fitting the observed profile with a synthetic profile of the Li doublet located at ~6708 Å. The synthetic spectrum was computed using the Turbospectrum code. Figure 5 shows four examples showing the method used to determine the A(Li).

thumbnail Fig. 5

Profile fitting of the Li doublet at ~6708 Å is shown for two UVES stars (CoRoT ID 101231832 and 102669801) and two Hydra stars (CoRoT ID 101124344 and 102586626). The observed spectra are presented using black lines, whereas the synthetic spectra are shown in red. The position of the Li doublet is indicated with a blue vertical line.

Open with DEXTER

The error σA(Li) in this abundance is related to the errors in the physical and chemical parameters; more specifically, the magnitude of this error is directly related to the error in Teff, and the errors in the other stellar parameters, such as log  (g), vmic, and vsin (i), produce minor effects in the measurements of A(Li). We determined new lithium abundances using synthetic spectra reflecting the errors in the four stellar parameters described above, which we called . Then, the final error σA(Li) is equal to the square root of the sum of the squares of the difference between A(Li) and , as is described in the following equation: (3)The Fig. 6 shows how the errors in the stellar parameters impact in the A(Li). In this context, the measurements of A(Li) for Hydra stars present errors higher than those found for the UVES stars. Nevertheless, we should be cautious with the Hydra data due, in particular, to the spectral resolution (R ~ 15 000) associated with the observations.

4. Results

The computed stellar parameters, including rotational velocities vsin (i) and lithium abundances, for the present stellar sample are listed in Table 7. Figure 7 presents the corresponding Hertzsprung-Russell (HR) diagram, with the stars segregated by their Galactic locations (CoRoT center/anticenter) and iron abundances. In the bottom panel, stars are divided in three different groups using their [Fe/H] values6. Errors in the parameters are also included in these panels. The magnitude of these errors is linked to the quality of the spectra, including spectral resolution R and S/N, and intrinsic effects of the stellar surfaces (i.e., high vsin (i), molecular bands in cool stars, etc.). Figure 7 shows that the present sample is comprised of stars in different evolutionary stages, ranging from the main sequence (MS) to the red giant branch (RGB).

thumbnail Fig. 6

As in Fig. 5, for the CoRoT ID 101231832 the profile fitting of an Fe I line and the Li doublet are presented. The panels a), b), c), and d) show how the errors in each stellar parameter impact the computed synthetic spectra, which were used to measure vsin (i) (panel e)) and A(Li) (panel f)), and their respective errors. For all panels, the observed spectrum is presented using a black line, whereas the synthetic spectrum, computed with the derived stellar parameters, is presented using a red line. The dashed green and blue lines represent the computed synthetic spectra with errors for a given stellar parameter.

Open with DEXTER

thumbnail Fig. 7

Upper panel: HR diagram of the stellar sample segregated by Galactic direction (center/anticenter). Blue and red dots represent stars in the LRc01 and LRa01 fields, respectively. Evolutionary tracks for Z = 0.019, from Girardi et al. (2000), are shown for stellar masses between 0.8 and 5 M. Lower panel: HR diagram of the stellar sample segregated by the iron abundance [Fe/H]. Green, black, and red points represent stars with [Fe/H]≤ − 0.25, −0.25<[Fe/H]≤ + 0.25 and [Fe/H]≥ + 0.25, respectively. Evolutionary tracks for Z = 0.004, Z = 0.019, and Z = 0.030, from Girardi et al. (2000), are presented with lines in green, black, and red, respectively. In this panel, only the following stellar mass values are represented: 0.8 M (dot lines), 1.0 M (solid lines) and 2.0 M (dash lines). For clarity, in both panels we plot only the evolutionary tracks until the RGB stage.

Open with DEXTER

We used the evolutionary tracks of Girardi et al. (2000) for different masses and metallicities7 to identify the evolutionary stages of the stars in our sample. For this purpose, we identified the turn-off and the base of the RGB from the evolutionary tracks for each Z to define the MS, subgiant branch (SGB), and RGB regions. The results of this classification are listed in Table 7. Mean values for the stellar parameters and their respectives standard deviations corresponding to different CoRoT fields and evolutionary stages are given in Table 8.

thumbnail Fig. 8

Upper panel: histogram of the surface gravities log  (g) in the CoRoT fields. Lower panel: histograms of the iron abundances [Fe/H] for the different evolutionary stages. A relation between [Fe/H] and evolutionary status is clearly present for the stars in our sample.

Open with DEXTER

Figure 7 also shows that the stellar evolutionary distribution for LRc01 and LRa01 CoRoT fields agrees with Deleuil et al. (2009). Furthermore, the relation between the color and evolutionary status is linked to the stellar metallicity. In fact, as we can see in Fig. 8 and Table 8, the distributions of log  (g) for both fields are different and these differences correlate with the [Fe/H] distributions. Most LRa01 stars have log  (g) 3.75 dex (~64%), whereas the LRc01 stars are spread along a large interval of log  (g) and most of them have log  (g) 3.75 dex. Combining this with the fact that the peaks of the [Fe/H] distributions and their mean values (Table 8) are different from one another, we can see that there is a relation between temperature, surface gravity, and evolutionary stage in the CoRoT fields considered here. While one might be tempted to associate these differences to selection effects, we note that these results agree with the recent spectroscopic survey of the CoRoT fields presented by G10. This point is discussed further in Sect. 4.4.

4.1. Comparison with previous results

To verify our results, we compared them with the results presented in G10. Only 11 stars of our sample were also analyzed by G10. These stars are listed in Table 9. In Fig. 9 we plot the comparison between our results and the G10 findings. Our results agree with the survey of G10, with Teff, log  (g) and vsin (i) presenting only small differences for most of the stars. However, star 101538522 presents a large dispersion in the log  (g) values, which may be explained because of the quality of the data of G10 for a RGB star (S/N ~ 23). We cannot directly compare their derived abundances with ours since G10 report only the global metallicity ([M/H]), whereas here we present the iron abundance ([Fe/H]).

Table 7

Stellar parameters for the stars in our sample.

4.2. Radial and rotational velocities

Figure 10 shows the distribution of the measurements of barycentric radial velocity vrad and rotational velocity vsin (i) listed in Tables 7 and 8. Small, but significant, differences are observed in the vrad distribution for stars located in the Galactic center and anticenter directions. The percentages of stars with vrad≤ 0km s-1 associated with the Galactic center and anticenter directions are 64% and 50%, respectively. In spite of the incompleteness of the sample, this behavior agrees with the vrad distribution determined by G10 (see their Fig. 2).

In Fig. 10, small differences are also observed in the distribution of vsin (i) values. In fact, 68% and 42% of stars in the LRc01 and LRa01 fields present vsin (i)≤ 5km s-1, respectively. The difference disapears quickly for stars with vsin (i)≤ 10km s-1, where the percentage for both fields are similar (89% and 83% for LRc01 and LRa01 fields, respectively). A few high rotation values can be noticed among stars in the Galactic center region. While we caution that these distributions are affected to some degree by incompleteness, the observed behavior of the vsin (i) distribution in particular does follow the behavior expected for FGK stars (Soderblom 1983; De Medeiros et al. 1996; Nordström et al. 2004). In fact, as we can see from Fig. 11, which displays the individual vsin (i) values versus Teff, the rotational behavior for stars of the present stellar sample is rather well in agreement with the well-established behavior of rotation for stars evolving from the MS to red giant stages. Essentially, stars in the MS exhibit a wide range of rotational velocity values, which are related with the stellar masses and Teff. For these stars, the measured values of vsin (i) ranges from a few km s-1 to about 100 times the solar rotation rate, whereas the stars along the RGB are typically slow rotators, except for a few unusual cases presenting moderate to rapid rotation (Cortés et al. 2009; Carney et al. 2008, 2003; De Medeiros et al. 1996; De Medeiros & Mayor 1990).

thumbnail Fig. 9

Comparison between our results and the results of G10 for stars in common. Some differences can be noted, however, overall our result agree with G10.

Open with DEXTER

Table 8

Mean parameters of the CoRoT fields.

Table 9

Common stars with Gazzano et al. (2010).

4.3. Photometric temperatures and reddening

It is possible to evaluate the reddening effects along the two different Galactic directions by comparing the initial photometric temperatures, derived without taking reddening into account, and the final, spectroscopically-derived, and presumably reddening-insensitive temperatures. This analysis allows one to establish how the determination of physical parameters from photometric data is affected by neglecting reddening effects, to evaluate the error budget brought about by reddening, and also to check the presence of possible reddening gradients in the CoRoT windows. In this sense, the present E(BV) estimates for individual stars may assist follow-up programs of specific groups of stars, including for instance solar analogs and solar twins.

To accomplish this goal, we compare the photometric temperatures Teff(BV) and Teff(JK) and those derived from our spectroscopic analysis, Teff. We determined the photometric temperatures, at the beginning, using the luminosity class from the CoRoTSky database. For some stars, however, the physical parameters provide a new classification of luminosity class for which new photometric temperatures were computed using this information. The photometric temperatures are listed in Table 7, including those derived using a new luminosity class. Finally, in Figs. 12 and 13, we present the comparisons between our spectroscopic Teff values and the photometric estimations Teff(BV) and Teff(JK).

Figure 12 shows that the stars in CoRoT run LRc01 present larger differences between Teff(BV) and Teff than those in run LRa01, which can be expected because of the higher extinction levels in the Galactic center direction. In fact, for this color index, (BV), the percentages of CoRoT stars presenting differences up to 200, 500, and 800 K in the Galactic center direction are 6%, 42%, and 81%, respectively. The percentages of CoRoT stars presenting the same temperature differences in the Galactic anticenter direction are 38%, 86%, and 92%, respectively. To obtain a reddening estimation for the LRc01 and LRa01 CoRoT fields, we used the calibration of Flower (1996; corrected by Torres 2010), which give us Teff(BV) and Teff, assuming the latter (spectroscopically derived) as being the actual value. This assumption is valid because Teff is not affected by reddening. As such, for a star with a solar value (BV), we estimated reddening levels (E(BV)) of about 0.06, 0.17, and 0.30, for the differences of 200, 500, and 800 K, respectively. Similarly, for an RGB star with a (BV)0 ~ 1.6, these differences in temperature represent reddening levels of about E(BV) ~ 0.07, 0.16, and 0.20, respectively.

thumbnail Fig. 10

Upper panel: histograms of the radial velocity vrad for the LRc01 (blue line) and LRa01 (red line) fields. Lower panel: histograms of the rotational velocity vsin (i) for the LRc01 (blue line) and LRa01 (red line) fields.

Open with DEXTER

thumbnail Fig. 11

Upper panel: distribution of vsin (i) for the stars in our sample along the HR diagram. Symbol size is proportional to the vsin (i) value. Evolutionary tracks are defined as in Fig. 7. Lower panels: histograms of the rotational velocity vsin (i) for the different evolutionary stages.

Open with DEXTER

thumbnail Fig. 12

Upper panel: differences between photometric Teff(BV) and spectroscopic Teff for stars in the LRc01 (open and filled blue dots) and LRa01 (filled and open red dots) CoRoT fields. Filled circles represent MS and SGB stars, whereas open circles represent RGB stars. Lower panels: histograms of the difference between photometric Teff(BV) and spectroscopic Teff for the stars in the LRc01 (blue line) and LRa01 (red line) CoRoT fields.

Open with DEXTER

thumbnail Fig. 13

Upper panel: differences between photometric Teff(JK) and spectroscopic Teff for the stars in the LRc01 (open and filled blue dots) and LRa01 (filled and open red dots) CoRoT fields. Filled circles represent MS and SGB stars, whereas open circles represent RGB stars. Lower panels: histograms of the difference between photometric Teff(JK) and spectroscopic Teff for the stars in the LRc01 (blue line) and LRa01 (red line) CoRoT fields.

Open with DEXTER

We also derived E(BV) values for individual targets using their values of Teff(BV) and Teff of Table 7. These values are given in Table 7. Table 8 shows the mean values E(BV) ⟩. Then, for all evolutionary stages, E(BV) ⟩ ~ 0.27 ± 0.14 and 0.12 ± 0.11 for the LRc01 and LRa01 fields, respectively. These mean values are somewhat influenced by the distributions of temperature and evolutionary stage in each field. Then we computed mean values E(BV) ⟩ for each evolutionary stage for both fields. For the LRc01 field, E ⟨ (BV) ⟩ levels are of 0.36 ± 0.22, 0.25 ± 0.11, and 0.25 ± 0.12 for MS, SGB and RGB stars, respectively, whereas for the LRa01 field, E(BV) ⟩ levels are of 0.11 ± 0.12, 0.12 ± 0.13, and 0.13 ± 0.06 for MS, SGB, and RGB stars, respectively.

On the other hand, there are also differences when the Teff(JK) values are compared with Teff in both CoRoT fields. The percentages of CoRoT stars presenting differences up to 200, 500, and 800 K in the Galactic center direction are 15%, 61%, and 90%, respectively. The percentages of CoRoT stars presenting the same temperature differences in the Galactic anticenter direction are 38%, 80%, and 91%, respectively.

The reddening levels for both CoRoT fields were determined using the relations of Alonso et al. (1996, 1999), which provide Teff(JK) and Teff, again assuming that the latter provides the correct value. As such, for a star with a solar value (JK), reddening levels are of about E(JK) ~0.04, 0.09, and 0.16 for differences up to 200, 500, and 800 K, respectively. Similarly, for a RGB star with a (JK) ~ 1.0 these differences in temperature imply a reddening of about E(JK) ~0.13, 0.37, and 0.71.

Similar to E(BV), we also derived E(JK) values for individual targets using their values of Teff(BV) and Teff of Table 7. These values are given in Table 7. Table 8 shows mean values E(JK) ⟩. Then, for all evolutionary stages, E(JK) ⟩ ~ 0.14 ± 0.10 and 0.10 ± 0.09 for the LRc01 and LRa01, respectively. For the LRc01 field, E(JK) ⟩ levels are of 0.10 ± 0.15, 0.11 ± 0.07, and 0.17 ± 0.08 for MS, SGB, and RGB stars, respectively, whereas for the LRa01 field E(JK) ⟩ levels are of 0.13 ± 0.10, 0.07 ± 0.09, and 0.14 ± 0.04 for MS, SGB, and RGB stars, respectively.

When we compare E(BV) ⟩ with E(JK) ⟩ for each field or the evolutionary stages in each field, typically they do not agree with one another. However, the dispersion in these mean values is very high, which could explain this discrepancy.

In contrast, the mean reddening values show that the LRc01 field is more affected by reddening than the LRa01 field. In fact, we used the reddening maps of Schlafly & Finkbeiner (2011) and Schlegel et al. (1998), and the Galactic Extinction Calculator8 of the NASA/IPAC Extragalactic Database (NED) to obtain mean values of reddening of the LRc01 and LRa01 fields. However, NED only provides reliable reddening values for the LRc01 field. The reddening values of Schlafly & Finkbeiner (2011) and Schlegel et al. (1998) agree with the present work. Specifically, for the LRc01 field Schlafly & Finkbeiner (2011) give reddening values E(BV) ~ 0.28 and E(JK) ~ 0.13, whereas Schlegel et al. (1998) give E(BV) ~ 0.30 and E(JK) ~ 0.16.

To make a comparison, we also obtained the color indices (BV) and (JK) for the stellar sample of G10. Again, we computed the Teff(BV) and Teff(JK) values for the stars for which those authors spectroscopically derived atmospheric parameters, and we obtained the differences between these temperatures and the spectroscopic temperatures. Histograms showing these differences for the stars in the LRc01 and LRa01 fields are presented in Fig. 14. There are some differences between these distribution and the corresponding histograms derived in Figs. 12 and 13. Compared to our sample, the G10 sample presents a greater proportion of highly-reddened stars in the Galactic center direction. In fact, for the (BV) color, the percentages of CoRoT stars presenting differences up to 200, 500, and 800 K in the Galactic center direction are 0%, 3%, and 30%, respectively. The percentages of CoRoT stars presenting the same temperature differences in the Galactic anticenter direction are 3%, 41%, and 80%, respectively.

This difference can probably be explained by the fact that the relative number of stars in the center and anticenter directions differs between the two studies. In addition, our sample size is only ~10% the size of the G10 sample, and sample size-related biases can also affect this comparison accordingly.

Otherwise, those stars with both determinations E(BV) and E(JK), only 27% present consistent values (E(BV) ~ 0.5 E(JK)), which can be produced by the errors in the photometry (see Sect. 3.3), and/or errors in determination of Teff and their respective errors. Therefore, it is important to take these values with caution.

thumbnail Fig. 14

Upper panel: histograms of the difference between photometric Teff(BV) and spectroscopic Teff for the sample of Gazzano et al. (2010). Lower panel: histograms of the difference between photometric Teff(JK) and spectroscopic Teff for the sample of Gazzano et al. (2010).

Open with DEXTER

thumbnail Fig. 15

Upper panel: histogram of the iron abundances [Fe/H] in the CoRoT fields. Lower panels: histograms of the iron abundances [Fe/H] for the different evolutionary stages. A relation between [Fe/H] and evolutionary status is clearly present for the stars in our sample.

Open with DEXTER

thumbnail Fig. 16

Upper panel: distribution of A(Li) along the HR diagram for the stars of the present sample. Symbol size is proportional to the value of A(Li). The stars showing an abnormal lithium behavior are presented using filled circles. Evolutionary tracks are the same as in Fig. 7. Lower panels: histograms of A(Li) for different evolutionary stages.

Open with DEXTER

4.4. The iron and Li abundances and three Li-rich giants

Figure 15 shows that our values of [Fe/H] strongly agree with typical values found in the Galactic disk, which range from −1.3 to + 0.4 (Reddy et al. 2006; Bensby et al. 2007; Meléndez et al. 2008). In fact, as was mentioned before, differences are expected in the distributions of [Fe/H] between both CoRoT fields. As shown in Fig. 2, the LRc01 field is composed of stars with lower [Fe/H] values than the LRa01 field. Stars with [Fe/H]< 0.0 represent 75% and 56% of the sample in the LRc01 and LRa01 fields, respectively. Following this point, stars in those CoRoT fields also present important differences in the distribution of Teff, log  (g), and evolutionary stages, suggesting a link with [Fe/H]. This can be seen when the histograms for each evolutionary stage are analyzed (see Fig. 15). The mean values of [Fe/H] for stars in the MS, SGB, and RGB of the LRc01 field are + 0.05, −0.16, and −0.26, respectively. For the LRa01 field, the corresponding mean values are + 0.04, −0.02, and −0.41.

Our [Fe/H] distributions show a similar behavior to that presented by Gazzano et al. (2010) in an extensive spectroscopic survey of the CoRoT field9. Similarly, our results show that most stars on the MS present solar [Fe/H] values, whereas most stars in evolved stages have low metallicities. The relation between [Fe/H], evolutionary stages, and the two different Galactic directions observed by CoRoT found in Gazzano et al. (2010), which we confirmed, could be explained by the metallicity gradient found in the Galactic disk (Pedicelli et al. 2009; Friel et al. 2010; Luck et al. 2011)10.

The behavior of the lithium abundances for the stars in our sample is shown in Fig. 16, with the A(Li) distribution along the HR diagram in the upper panel and histograms for different evolutionary stages in the lower panel. The A(Li) measurements for the present stellar sample clearly follows the well-established scenario for the lithium behavior at the referred evolutionary phases (Luck 1977; Boesgaard & Tripicco 1986; Soderblom et al. 1993; Wallerstein et al. 1994; De Medeiros et al. 1997; Lèbre et al. 1999; De Medeiros et al. 2000; Meléndez et al. 2010).

Indeed, the stellar lithium content is extremely sensitive to the physical conditions inside stars. As well established, the surface Li abundance is further depleted after stars leave the MS and undergo the first dredge-up (Iben 1967a,b). As a result, RGB stars essentially exhibit low A(Li) (Brown et al. 1989). Nevertheless, an increasing list of studies report the discovery of giant stars violating this rule (e.g., Wallerstein & Sneden 1982; Brown et al. 1989; Pilachowski et al. 2000; Martell & Shetrone 2013), the so-called lithium-rich giant stars, which present atypically large lithium abundances, in contrast to theoretical predictions. Three stars in the present CoRoT sample show an abnormal lithium behavior: CoRoT ID 100537408, with an A(Li) of 1.45 ± 0.46; CoRoT ID 101358013, with an A(Li) of 2.27 ± 0.21; and CoRoT ID 101555541, with an A(Li) of 1.13 ± 0.18 (see Fig. 16).

Figure 17 shows the distributions of A(Li) for the whole sample located in the Galactic center and anti-center directions. The upper panel of Fig. 17 provides an indication that a difference may be present, with the highest lithium content in the anti-center direction and an apparent bimodal distribution in the center direction. However, the histograms for the stars segregated by MS, SGB, and RGB evolutionary stages (bottom panel) show no statistically clear difference between Galactic center and anti-center. Indeed, the distributions of A(Li) for stars separated by evolutionary stages seem to indicate that the behavior of the lithium content observed along the HR diagram follows the same trend irrespective of the location of stars in Galactic center or anti-center directions.

thumbnail Fig. 17

Upper panel: histogram of the lithium abundances A(Li) in the CoRoT fields with stars segregated in the two Galactic directions, with blue indicating the Center and red the Anti-center. Lower panels: histograms of A(Li) for different evolutionary stages.

Open with DEXTER

5. Conclusions

We present physical parameters (Teff, log  (g), vmic, and vsin (i)), dynamical properties (vrad), and chemical abundances ([Fe/H] and A(Li)) for a sample of 138 stars stars located in the CoRoT exoplanet fields LRc01 and LRa01, based on observations collected with UVES/VLT at ESO and Hydra/Blanco at CTIO. The derived parameters allow us to characterize physically the observed stellar sample, with the stars located in different regions of the HR diagram, from the MS to well-evolved stages, including the SGB and RGB. We also provide estimates of possible errors in the temperatures derived using photometric calibrations, and also of reddening values for the stars in the aforementioned CoRoT fields.

Our results show a relation between Teff, evolutionary stage, and [Fe/H] in the CoRoT fields, which is related to the color distributions in these fields. These results are in agreement with independent photometric and spectroscopic surveys of the CoRoT fields. These results give support to our spectroscopically-determined parameters. Our chemical analysis shows that the stars in the CoRoT fields present the same patterns found and reported on for the Galactic disk, showing a mixture of different populations associated with the Milky Way.

The stellar sample presents the same rotational behavior described in the literature for different evolutionary stages and colors. Also, we provide a calibration to derive vsin (i) from Hydra observations using the robust CCF technique (see Sect. 3.2 and Eq. (3)).

Finally, the present data set also represents an important piece of work to be used as standard sample calibration for different programs in the context of the CoRoT mission, since, among the brightest stars that comprise the CoRoT exoplanet field targets, dozens are included in the list of stars analyzed here. This work can also help to increase the scientific return of other spacial missions, such as Gaia or TESS.


2

IRAF is distributed by the National Optical Astronomy Observatories, which are operated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation.

3

These tables are available on the Kurucz webpage http://kurucz.harvard.edu/

4

We use the Fe I and Fe II abundances defined as and , respectively.

5

Here the Li abundance is defined as .

6

[Fe/H] is calculated as in Canto Martins et al. (2011) using a solar iron abundance A(Fe) = 7.49 dex.

7

For the different [Fe/H] groups, we used evolutionary tracks with a representative Z. Specifically, we used metallic abundances Z = 0.004, 0.019 and 0.030 for the group with [Fe/H]≤ − 0.25, −0.25 <[Fe/H]≤ + 0.25 and [Fe/H]≥ + 0.25, respectively.

8

Available in the NED web page http://ned.ipac.caltech.edu/forms/calculator.html

9

We present measurements of [Fe/H], whereas G10 present the global metallicity [M/H]. The comparison should be done with caution since both quantities do not represent the same abundance.

10

Our stellar sample comprises stars in early and evolved stages and they present a narrow interval in apparent magnitudes V, which implies a spread in absolute magnitudes and distances.

Acknowledgments

Research activities of the Stellar Board of the Federal University of Rio Grande do Norte are supported by continuous grants of CNPq and FAPERN Brazilian agencies. We also acknowledges financial support of the INCT INEspaço. I.C.L. and C.E.F.L. acknowledges postdoctoral fellowship of the CNPq; C.C., S.C.M., C.E.F.L., S.V. and G.P.O. acknowledge graduate fellowships of the CAPES agency. This work was partially supported by the German Deutsche Forschungsgemeinschaft, DFG project number Ts 17/201. Support for C.C. and M.C. is provided by the Chilean Ministry for the Economy, Development, and Tourism’s Programa Iniciativa Científica Milenio through grant IC 120009, awarded to the Millennium Institute of Astrophysics (MAS); by Proyecto Basal PFB- 06/2007; and by Proyecto FONDECYT Regular #1141141. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

References

All Tables

Table 1

CoRoT IDs, ephemerides, details of observations, photometry, and estimated luminosity classes (LC) for the UVES stars.

Table 2

Setup used for Hydra observations.

Table 3

Main characteristics for the Hydra stars.

Table 4

Main characteristics and rotation velocities vsin (i) of calibrator stars.

Table 5

Observed broadening σobs and σt for template stars.

Table 6

Observed broadening σobs and σobs0 for calibrator stars.

Table 7

Stellar parameters for the stars in our sample.

Table 8

Mean parameters of the CoRoT fields.

All Figures

thumbnail Fig. 1

Some spectra of the stars contained in the sample. Spectra collected using the UVES spectrograph are show in the upper panel, whereas the lower panel shows spectra collected using the Hydra spectrograph.

Open with DEXTER
In the text
thumbnail Fig. 2

Comparison between the vrad measurements for the Hydra sample obtained using different synthetic spectra. The represents the difference between the vrad obtained when the Hydra spectra are cross-correlated with synthetic spectra for a RGB star and the Sun, respectively. The red line represents the mean value of and the standard deviation is presented using black dashed lines. Also, the typical error in vrad measurements for Hydra stars is shown using an error bar.

Open with DEXTER
In the text
thumbnail Fig. 3

CCF-vsin (i) calibration for the Hydra spectrograph. The dashed line represents the function found and black points represent the values for the calibrator stars, respectively.

Open with DEXTER
In the text
thumbnail Fig. 4

Upper panels: ionization equilibrium for the targets 100932329 (UVES star) and 101565091 (Hydra star). Lower panels: equilibrium of the A(Fe)s and their EWs. Open and filled circles represent the abundances of Fe I (A(FeI)) and Fe II (A(FeII)), respectively. Red and black dashed lines represent the mean value of A(Fe) and standard deviation, respectively.

Open with DEXTER
In the text
thumbnail Fig. 5

Profile fitting of the Li doublet at ~6708 Å is shown for two UVES stars (CoRoT ID 101231832 and 102669801) and two Hydra stars (CoRoT ID 101124344 and 102586626). The observed spectra are presented using black lines, whereas the synthetic spectra are shown in red. The position of the Li doublet is indicated with a blue vertical line.

Open with DEXTER
In the text
thumbnail Fig. 6

As in Fig. 5, for the CoRoT ID 101231832 the profile fitting of an Fe I line and the Li doublet are presented. The panels a), b), c), and d) show how the errors in each stellar parameter impact the computed synthetic spectra, which were used to measure vsin (i) (panel e)) and A(Li) (panel f)), and their respective errors. For all panels, the observed spectrum is presented using a black line, whereas the synthetic spectrum, computed with the derived stellar parameters, is presented using a red line. The dashed green and blue lines represent the computed synthetic spectra with errors for a given stellar parameter.

Open with DEXTER
In the text
thumbnail Fig. 7

Upper panel: HR diagram of the stellar sample segregated by Galactic direction (center/anticenter). Blue and red dots represent stars in the LRc01 and LRa01 fields, respectively. Evolutionary tracks for Z = 0.019, from Girardi et al. (2000), are shown for stellar masses between 0.8 and 5 M. Lower panel: HR diagram of the stellar sample segregated by the iron abundance [Fe/H]. Green, black, and red points represent stars with [Fe/H]≤ − 0.25, −0.25<[Fe/H]≤ + 0.25 and [Fe/H]≥ + 0.25, respectively. Evolutionary tracks for Z = 0.004, Z = 0.019, and Z = 0.030, from Girardi et al. (2000), are presented with lines in green, black, and red, respectively. In this panel, only the following stellar mass values are represented: 0.8 M (dot lines), 1.0 M (solid lines) and 2.0 M (dash lines). For clarity, in both panels we plot only the evolutionary tracks until the RGB stage.

Open with DEXTER
In the text
thumbnail Fig. 8

Upper panel: histogram of the surface gravities log  (g) in the CoRoT fields. Lower panel: histograms of the iron abundances [Fe/H] for the different evolutionary stages. A relation between [Fe/H] and evolutionary status is clearly present for the stars in our sample.

Open with DEXTER
In the text
thumbnail Fig. 9

Comparison between our results and the results of G10 for stars in common. Some differences can be noted, however, overall our result agree with G10.

Open with DEXTER
In the text
thumbnail Fig. 10

Upper panel: histograms of the radial velocity vrad for the LRc01 (blue line) and LRa01 (red line) fields. Lower panel: histograms of the rotational velocity vsin (i) for the LRc01 (blue line) and LRa01 (red line) fields.

Open with DEXTER
In the text
thumbnail Fig. 11

Upper panel: distribution of vsin (i) for the stars in our sample along the HR diagram. Symbol size is proportional to the vsin (i) value. Evolutionary tracks are defined as in Fig. 7. Lower panels: histograms of the rotational velocity vsin (i) for the different evolutionary stages.

Open with DEXTER
In the text
thumbnail Fig. 12

Upper panel: differences between photometric Teff(BV) and spectroscopic Teff for stars in the LRc01 (open and filled blue dots) and LRa01 (filled and open red dots) CoRoT fields. Filled circles represent MS and SGB stars, whereas open circles represent RGB stars. Lower panels: histograms of the difference between photometric Teff(BV) and spectroscopic Teff for the stars in the LRc01 (blue line) and LRa01 (red line) CoRoT fields.

Open with DEXTER
In the text
thumbnail Fig. 13

Upper panel: differences between photometric Teff(JK) and spectroscopic Teff for the stars in the LRc01 (open and filled blue dots) and LRa01 (filled and open red dots) CoRoT fields. Filled circles represent MS and SGB stars, whereas open circles represent RGB stars. Lower panels: histograms of the difference between photometric Teff(JK) and spectroscopic Teff for the stars in the LRc01 (blue line) and LRa01 (red line) CoRoT fields.

Open with DEXTER
In the text
thumbnail Fig. 14

Upper panel: histograms of the difference between photometric Teff(BV) and spectroscopic Teff for the sample of Gazzano et al. (2010). Lower panel: histograms of the difference between photometric Teff(JK) and spectroscopic Teff for the sample of Gazzano et al. (2010).

Open with DEXTER
In the text
thumbnail Fig. 15

Upper panel: histogram of the iron abundances [Fe/H] in the CoRoT fields. Lower panels: histograms of the iron abundances [Fe/H] for the different evolutionary stages. A relation between [Fe/H] and evolutionary status is clearly present for the stars in our sample.

Open with DEXTER
In the text
thumbnail Fig. 16

Upper panel: distribution of A(Li) along the HR diagram for the stars of the present sample. Symbol size is proportional to the value of A(Li). The stars showing an abnormal lithium behavior are presented using filled circles. Evolutionary tracks are the same as in Fig. 7. Lower panels: histograms of A(Li) for different evolutionary stages.

Open with DEXTER
In the text
thumbnail Fig. 17

Upper panel: histogram of the lithium abundances A(Li) in the CoRoT fields with stars segregated in the two Galactic directions, with blue indicating the Center and red the Anti-center. Lower panels: histograms of A(Li) for different evolutionary stages.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.