Spectroscopy of Dwarf Stars Around the North Celestial Pole
Šarūnas Mikolaitis, Gražina Tautvaišienė, Arnas Drazdauskas, Renata Minkevičiūtė, Lukas Klebonas, Vilius Bagdonas, Erika Pakšienė, Rimvydas Janulis
DDraft version January 7, 2019
Typeset using L A TEX twocolumn style in AASTeX61
SPECTROSCOPY OF DWARF STARS AROUND THE NORTH CELESTIAL POLE ∗ ˇSar¯unas Mikolaitis, Graˇzina Tautvaiˇsien ˙e, Arnas Drazdauskas, Renata Minkeviˇci¯ut ˙e, Lukas Klebonas,
1, 2
Vilius Bagdonas, Erika Pakˇsien ˙e, and Rimvydas Janulis Institute of Theoretical Physics and Astronomy, Vilnius University, Saul˙etekio av. 3, 10257 Vilnius, Lithuania Mathematisch-Naturwissenschaftliche Fakult¨at, Universit¨at Bonn, Wegelerstraße 10, 53115 Bonn, Germany (Received 2018 March 1; Accepted 2018 April 1)
Submitted to PASPABSTRACTNew space missions (e.g. NASA-TESS and ESA-PLATO) will perform an in-depth analysis of bright stars in largefields of the celestial sphere searching for extraterrestrial planets and investigating their host-stars. Asteroseismicobservations will search for exoplanet-hosting stars with solar-like oscillations. In order to achieve all the goals, a fullcharacterization of the stellar objects is important. However, accurate atmospheric parameters are available for lessthan 30% of bright dwarf stars of the Solar neighborhood. In this study we observed high-resolution (R=60000) spectrafor all bright (
V < α (2000) = 161.03 ◦ and δ (2000) = 86.60 ◦ that is a centre of one of thepreliminary ESO-PLATO fields. Spectroscopic atmospheric parameters were determined for 140 slowly rotating stars,for 73% of them for the first time. The majority (83%) of the investigated stars are in the TESS object lists and all ofthem are in the preliminary PLATO field. Our results have no systematic differences when compared with other recentstudies. We have 119 stars in common with the Geneva-Copenhagen Survey, where stellar parameters were determinedphotometrically, and find a 14 ±
125 K difference in effective temperatures, 0 . ± .
16 in log g , and − . ± .
09 dexin metallicities. Comparing our results for 39 stars with previous high-resolution spectral determinations, we find onlya 7 ±
73 K difference in effective temperatures, 0 . ± .
09 in log g , and − . ± .
09 dex in metallicities. We alsodetermined basic kinematic and orbital parameters for this sample of stars. From the kinematical point of view, almostall our stars belong to the thin disk substructure of the Milky Way. The derived galactocentric metallicity gradientis − . ± .
024 dex kpc − (2.5 σ significance) and the vertical metallicity gradient is − . ± .
099 dex kpc − (1 σ significance) that comply with the latest inside-out thin disk formation models, including those with stellar migrationtaken into account. Keywords:
Galaxy: solar neighborhood – stars: atmospheres
Corresponding author: ˇSar¯unas [email protected] ∗ Based on observations collected with the 1.65 m telescope and VUES spectrograph at the Mol˙etai Astronomical Observatory of Instituteof Theoretical Physics and Astronomy, Vilnius University, for the SPFOT survey. a r X i v : . [ a s t r o - ph . S R ] J a n Mikolaitis et al. INTRODUCTIONStars that are in a close proximity to our Solar systemare the ones that can be studied in detail using goodquality of astrometric, photometric and spectroscopicmeasurements. In the era of new planetary search spacemissions (e.g. NASA-TESS or ESA-PLATO) brightstars in the Solar neighborhood become even more im-portant to investigate. If a planetary system is foundaround a bright nearby star it gets a large attentionfrom many astrophysical perspectives. A full astero-seismic characterisation of a star is possible if its mainatmospheric parameters are known.It could be thought that cool bright main sequencestars in the Solar neighborhood should have been care-fully studied, however, only about 30% of F5 and coolerdwarfs with magnitudes
V < V =12 mag.Large telescopes cannot pay much attention in sur-veying nearby bright stars, leaving the study of Solarneighborhood mainly to archival data (cf. Adibekyanet al. 2012, Worley et al. 2012, 2016; De Pascale et al.2014), however all these samples cover only a small frac-tion of the sky. Such spectroscopic surveys of stars inthe Solar neighborhood like Allende Prieto et al. (2004)who studied in a high-resolution mode all stars up to V =6.5 mag around the Sun (nearest 15 pc) are veryrare.The metllicity gradients of the disk are importanttools in studying the Galactic disk formation. Usu-ally, this task lies on large spectroscopic surveys (e.g.Co¸skunoˇglu et al. 2012; Boeche et al. 2013; Cheng et al.2012; Duong et al. 2018). The Solar proximity stars are not well spread to provide meaningful vertical or radialgradients. However, kinematic and orbital computa-tions show the statistical positions of these stars duringtheir movement around the Galaxy. The orbital param-eters such as the mean galactocentric distance show theoriginal spatial distribution of the presently neighboringstars and can be studied in the context of metallicitygradients (e.g. Anders et al. 2014; Boeche et al. 2013).Therefore, bright ( V <
V < TARGET SELECTIONAn initial list of 1163 stars was compiled from theTYCHO-2 catalog (Høg et al. 2000) as follows. Wesearched the catalogue considering two main criteria.Firstly, we approximated the shape of the field to adisk with the radius of 20 degrees around the centreof the field. Northern celestial pole is very close tothe centre of one of the possible PLATO fields (Raueret al. 2014, 2016) thus we centered our observations ac-cordingly (20 degrees around α (2000) = 161.03552 ◦ and pectroscopy of dwarf stars around the north celestial pole - 0 . 4 0 . 0 0 . 4 0 . 8 1 . 2 1 . 6 2 . 050- 5 M V ( B (cid:1) V ) Figure 1.
Color-magnitude diagram of stars in the inves-tigated field. The FGK dwarfs observed in this programmeare presented as red open circles.
Figure 2.
Postitions (RA and DEC in degrees) of the pro-gramme stars (red open circles). The centre of the field isshown as the black cross. δ (2000) = 86.60225 ◦ ). Secondly, we limited a number ofstars by setting the visual magnitude V < B − V ) and M V . Due to the large size of our field, it was difficultto investigate the extinction law abnormalities. Thus,according to the mean extinction law we adopted the R = 3 . E B − V values were calculatedusing the model of large-scale visual interstellar extinc-tion by Hakkila et al. (1997). The parallaxes requiredfor the stellar distance calculations were gathered fromthe HIPPARCOS Catalog (Perryman et al. 1997). Wehave selected to observe 6500 K and cooler main se-quence and sub-giant stars. It complies with the obser-vational strategies of TESS and PLATO. The majorityof the stars in the Solar neighborhood are more metalrich than − . / H] = − . T eff ≈ B − V ) ≈ .
39. Therefore,we have constructed a color-magnitude diagram for allselected 1163 stars in the field (see Figure 1). We choseonly dwarf and subgiant stars with ( B − V ) > OBSERVATIONSThe programme stars were observed with the VilniusUniversity Echelle Spectrograph (VUES) designed andconstructed at the Exoplanet Laboratory of the YaleUniversity (Jurgenson et al. 2014, 2016) and mountedon the f/12 1.65 meter Ritchey-Chretien telescope atthe Mol˙etai Astronomical Observatory of the Instituteof Theoretical Physics and Astronomy, Vilnius Univer-sity. The VUES is designed to observe spectra in the4 000 to 8 800 ˚A wavelength range with three spectralresolution modes ( R = 30 000, 45 000, and 60 000). Fig-ure 3 shows an example of solar spectra obtained withthe VUES ( R = 60 000) and several other well knownspectrographs (upper panel), the lower panel shows fluxdifferences between the VUES and other spectrographs.During an observational period of 2016–2017 we ob-tained 365 spectra of 213 stars. For stars that wereobserved several times, their spectra were combined inorder to increase a signal-to-noise ratio. A sample ofthe observation log is provided in Table 1. Bias, flatfield, and calibration lamp measurements were acquiredevery evening before observing stellar spectra. We useda quartz lamp for the flat fielding and the ThAr spectrafor the wavelength calibration. The data were reducedand calibrated following standard reduction procedures Mikolaitis et al. N o r m a li s e d f l u x VUESATLASNARVALHARPS λ , Å ∆ f l u x VUES − ATLASVUES − NARVALVUES − HARPS
Figure 3.
Upper panel: examples of the Solar spectra from the ATLAS (Wallace et al. 2011), HARPS (Mayor et al. 2003),NARVAL (Auri`ere 2003), and VUES (Jurgenson et al. 2016) spectrographs (original resolutions). Lower panel: the fluxdifferences between VUES and other spectra (the comparison spectra were downgraded to match the resolving power of VUES).
Table 1.
Sample table of a summary of observations.Tycho2 ID TESS ID (TIC-5) Date and UTC time Exposure time (s) RA* (h:m:s) DEC* ( ◦ : (cid:48) : ”) S/NTYC 4634-2068-1 288183829 2016-06-23 01:35:49.755 1200 14:48:6.16 +82:27:4.565 87TYC 4634-2068-1 288183829 2017-02-07 04:54:41.028 2400 14:49:5.702 +82:25:23.522 87TYC 4615-1144-1 401497601 2016-08-25 03:14:50.853 2400 01:01:0.882 +83:11:2.74 48TYC 4615-1144-1 401497601 2016-09-07 03:37:47.693 2400 01:00:57.999 +83:11:0.075 48TYC 4615-1144-1 401497601 2016-10-18 00:09:01.201 2400 01:01:5.135 +83:11:23.529 48Notes. A full table is only available in an electronic form at the CDS.Coordinates correspond to the star position at the time of the observation. Table 2.
Sample table of the kinematic properties of the observed stars.Tycho2 ID U LRS V LRS W LRS R mean z max e T D/D V rad σ V rad FWHM V rad TYC 4634-2068-1 − .
70 23.45 19.42 9.14 0.37 0.24 0.11 − . − .
92 7.81 0.23 0.04 0.01 − . which included a subtraction of the bias frame, correc-tion for flat field, extraction of orders, wavelength cal-ibration, and a cosmic rays removal (Jurgenson et al.2016). METHOD OF ANALYSIS4.1.
Radial velocity determination and identification ofdouble-line binaries and fast-rotating stars
Since the target selection was done using only thephotometric indices, a number of fast-rotating stars orspectroscopic double-line binaries were expected in thetarget list. Therefore, we performed a simple cross-correlation of spectra with a mask based on the atomicline list in order to detect double-line binaries andcompute radial velocities. The line list for the maskwas combined by the Gaia-ESO survey line list group pectroscopy of dwarf stars around the north celestial pole −
300 to +300 km s − was calculated in radial veloc-ity steps of ∆V rad =1.2 km s − . Figure 4 shows theCCF corresponding to the fit for three spectra: a slowrotating single-line star, a fast rotator and a double-line binary. After the visual inspection, we have omit-ted 15 stars that have double-line features in theirspectra. Eight of them are known spectroscopic bina-ries: HD 223778, HD 94686, HD 166865, HD 110533,HD 166866, HD 205234, HD 195850, HD 209942 (Pour-baix et al. 2004; Nordstr¨om et al. 2004). For other starsGaussian fits were made in order to determine a min-imum of the profile, and hence the radial velocity asit is shown in Figure 5. Figure 6 is a histogram ofthe radial velocities calculated for the 213 stars. Themajority of the spectra have radial velocities between −
40 and +20 km s − .Within our dataset, we found 139 stars in commonwith the Nordstr¨om et al. (2004) study. Their radial ve-locity data were obtained with the photoelectric crosscorrelation spectrometers CORAVEL (Baranne et al.1979; Mayor 1985). In Figure 7, we show a compari-son of the Nordstr¨om et al. (2004) radial velocity val-ues with those determined in our study. The mean andstandard deviation of differences between the two setsis (cid:104) ∆ V rad (cid:105) = 0 . ± . − . Figure 8 displays ahistogram of the FWHM of the CCF (the red dottedline in Figure 5). It shows that the majority of spectrareturned a FWHM of less than 20 km s − . This FWHMindirectly reflects the rotational velocity of a star thatbroadens the observed spectral features depending onthe value of the rotational velocity. It was not possibleto measure equivalent widths of lines for 51 stars witha satisfactory quality because of the broad and blendedlines. All these fast-rotating stars show the FWHM ofthe CCF larger than 25 km s − . Nordstr¨om et al. (2004)have derived v sin i for 28 of our 57 fast-rotating starsand for 96 of our 140 slow-rotating stars. Based on thatcatalog, 28 of our fast-rotating stars rotate from 19 to70 km s − , while other 96 stars rotate up to 15 km s − .Therefore we have excluded 16 stars that have double-line features in their spectra and 57 fast-rotating stars.A comprehensive analysis of these stars will be carriedout in a separate work. The final sample consists of140 slow-rotating stars, and the atmospheric parametersare provided in this paper for all of them.4.2. Kinematic properties
Kinematic values for our stars were calculated usingthe python based package for galactic-dynamics calcu- ca CC F b CC F CC F Vrad, km s-1
Figure 4.
CCFs produced for calculating the radial ve-locities and detection of double-line binary stars: a) atypical slow-rotating star, b) a fast-rotating star, c) CCFof the double-line spectroscopic binary TYC 4602-552-1(HIP 117712), showing two profiles. - 1 0 0 1 0 2 0 3 0 4 00 . 9 00 . 9 51 . 0 0
V r a d k m s - 1
CCF
Figure 5.
The same CCF function from Figure 4a with thefitted Gaussian function (red line), from which the radialvelocities were determined. The FWHM of the Gaussian isshown as the red dotted line. lations
Galpy by Bovy (2015). A sample of kinematicparameters is presented in Table 2. Paralaxes, proper http://github.com/jobovy/galpy Mikolaitis et al. -140 -120 -100 -80 -60 -40 -20 0 20 40 6001020304050607080 N u m be r o f s t a r s Vrad, km s-1
Figure 6.
Histogram of the radial velocity values calculatedfor all targeted stars. -80 -60 -40 -20 0 20 40 60-80-60-40-200204060 〈 N04 V rad - V rad 〉 = -0.52 σ = N V r ad , k m s - VUES Vrad, km s-1
Figure 7.
Comparison of radial velocities derived in thisstudy and by Nordstr¨om et al. (2004) (139 stars). The reddashed line with a slope of 1 is shown for comparison. motions and coordinates required for the space veloci-ties and galactic position calculations were taken fromthe TGAS (
The Tycho-Gaia Astrometric Solution ) cat-alogue (Michalik et al. 2015). The radial velocities ofprogramme stars were determined in this work. To re-late the calculated parameters to the Sun, we have usedthe solar distance from the Galactic plane z (cid:12) = 0 .
02 kpc(Joshi 2007). The Sun’s motion components relativeto the local standart of rest ( U (cid:12) , V (cid:12) , W (cid:12) ) = (11.10,12.24, 7.25) km s − were adopted from Sch¨onrich et al.(2010). The orbital parameters were calculated usingthe default potential ( MWPotential2014 ) consisting ofa bulge, disk and NFW halo (see Bovy 2015) and the in-tegration time of 5 Gyrs. The rotational velocity of the N u m be r o f s t a r s FWHM of Vrad, km s-1
Figure 8.
Histogram of the FWHM of the V rad values cal-culated for all targeted stars. disk at the Sun’s Galactocentric radius in MWPoten-tial2014 is fixed to V c =220 km s − that is adopted fromthe APOGEE spectroscopic survey data by Bovy et al.2012 ( V c =218 ± − ). The value V c =220 km s − is also recommended by the International Astronomi-cal Union (IAU) (Kerr & Lynden-Bell 1986). However,there are still debates whether it needs to be revised up-ward to around V c =236 km s − (Kawata et al. 2018),to V c =240 km s − (Reid & Dame 2016), or even up to V c =250 km s − and more (Reid et al. 2009; Sch¨onrich2012).4.3. Determination of atmospheric parameters
We used a pipeline of analysis that was constantlyused for the Gaia-ESO survey computations by the Vil-nius node. It is described in Smiljanic et al. (2014), herewe note the most important information.pa Stellar atmospheric parameters were determinedusing traditional equivalent width (EW) based meth-ods. EWs were measured using the DAOSPEC (Stetson& Pancino 2008) software. Figure 9 show a compari-son of EWs measured for the solar Fe I lines from theVUES spectra and the HARPS and NARVAL spectra,respectively. The EWs measured from spectra of thesespectrographs agree very well.Effective temperatures were determined by minimiz-ing a slope of abundances obtained from Fe I lines withrespect to the excitation potential. Surface gravitieswere determined by forcing the measured Fe I and Fe IIlines to yield the same iron abundance. Microturbulentvelocities were determined by forcing Fe I abundancesto be independent of the EWs of the lines. A customwrapper software was developed to measure EWs, and pectroscopy of dwarf stars around the north celestial pole 〈 �� ������ �� �� ���� 〉 �� ����� σ = ���� 〈 �� ����� �� �� ���� 〉 �� ������ σ = ���� � � � � � � �� � � � � � �� ���� �� �� � � � � � �� � � � � Figure 9.
Comparison of equivalent widths of solar ironlines measured in the HARPS, NARVAL and VUES spec-tra. The red dashed lines with a slope of 1 are shown forcomparison. compute the main atmospheric parameters and abun-dances automatically.The Fe I and Fe II linelist (299 lines) was taken fromSousa et al. (2008) with . The stellar atmospheric pa-rameters were computed using the 10-th version of theMOOG code (Sneden 1973) using a grid of MARCS stel-lar atmosphere models (Gustafsson et al. 2008). The fi-nal interpolated model in the WEBMARCS format forMOOG was calculated using a modified interpolationsoftware, provided together with the MARCS models.The pipeline performs an iterative sequence of abun-dance calculations in order to minimize the abundancedependency on the line excitation potential, [Fe I/Fe II]and σ ([Fe I/H]). It is done by employing the Nelder-Mead method (Nelder & Mead 1965). Every resultingabundance for every single line that departed from themean value by more than 2 σ was flagged as an outlier.4.4. Error estimation for the atmospheric parameters
Using these previously described techniques, we alsoestimated the different sources of possible uncertaintiesraising in deriving stellar atmospheric parameters. Typ-ically there are multiple sources of uncertainties, someof which affect single lines independently (e.g. randomerrors of the line fitting or continuum placement) as wellas errors of the employed method, which are mainly as-sociated with the uncertain linear regression fit or theatomic parameters.First, we studied the errors that might be caused by apossible incorrect continuum placement. There are sev-eral ways to evaluate these errors. One is to perform theMonte Carlo simulations, selecting a statistically signif-icant set of spectra and adding a noise artificially. Thissimulation can show the sensitivity of the method tonoises and continuum placement. Another way is to fol-low the line-to-line scatter.If there is a statistically significant number of lines fora given element, the scatter informs about the combinedeffect of the erroneous continuum placements, equivalentwidth measurements, and uncertain atomic parametersfor different lines.For the Monte Carlo simulations, we took spec-tra of two stars that are most typical for the sam-ple: TYC 4573-1916-1 ( T eff =6153 K, log g =4.01,[Fe/H]= − .
07) and the Sun ( T eff =5779 K, log g =4.49,[Fe/H]= − . . The spectra of these bright stars havea signal-to-noise ratio (SNR) of more than 200 per pixel.In order to investigate a possible noise influence, we de-graded these spectra with a white Gaussian noise tothe SNR equal to 25, 50, and 75 per pixel. We thengenerated 100 spectra for each SNR value, remeasuredequivalent widths and derived the corresponding 100atmospheric parameters for each SNR. In that way, wedetermined sensitivity of our atmospheric parametersto the SNR. These uncertainties are provided in a formof the standard deviations in Table 3.However, this kind of error estimation is only robustwhen there are many lines. We were able to use 92–252Fe I and 7–32 Fe II lines in the analysis. The line-to-line We note that the Solar values of our method ( T eff =5779 ±
24 K,log g =4.49 ± .
12, [Fe/H]= − . ± T eff =5772.0 K, log g = 4 . T eff =5771 K, log g =4.4380, Heiter et al.2015b). Both Heiter et al. (2015b) and Prˇsa et al. (2016) studiesbenefit from very precise methods that are independent of spec-troscopy and atmospheric models and reach errors in T eff and log g up to 1 K and 0.0002 cm s − , respectively. Similarly, the metal-licity value could be slightly off 0.0 dex in different methods (e.g.Table 2 in Jofr´e et al. 2014). Mikolaitis et al.
Table 3.
Errors due to the uncertain continuum placementand equivalent width measurement. Based on the MonteCarlo simulations.S/N=25 S/N=50 S/N=75TYC 4573-1916-1 T eff = 6153 K, log g = 4 .
01, [Fe / H] = − . σ T eff
37 31 28 σ log g σ [Fe / H] σ v t ∗ T eff = 5779 K, log g = 4 .
49, [Fe / H] = − . σ T eff
52 36 35 σ log g σ [Fe / H] σ v t ∗ Solar atmospheric parameters derived with our method(see Section 4.4). l og g T eff , K -0.5-0.4-0.3-0.2-0.100.10.30.4 Figure 10.
Temperature–gravity diagram of investigatedstars (dots) with metallicity coded in color. Evolutionarytracks by Girardi et al. (2000) with masses between 0.7 and1.9 M (cid:12) and Z ini =0.019 are plotted by the grey solid lines. scatter of Fe I and Fe II abundances was propagated tothe uncertainties of atmospheric parameters as follows: • The uncertainty for the effective temperatures wasestimated by obtaining the boundary temperaturevalues of the possible satisfactory parameter space, using the error of the linear regression fit to theFe I abundances. • The uncertainty of the surface gravity was ob-tained searching for the boundary values of log g according to the standard deviations of abun-dances from the Fe I and Fe II lines. • The uncertainty of the microturbulent velocity wasobtained by employing the error of the standarddeviation of the neutral iron abundances. • The [Fe/H] standard deviation ( σ [Fe I/H]) wasadopted as the metallicity uncertainty.These uncertainties are provided for every star in Table 4as well. STELLAR PARAMETERS5.1.
Atmospheric parameters
We have determined values of effective temperature,surface gravity, metallicity, microturbulent velocity, andradial velocity with their corresponding uncertaintiesfor 140 stars and the sample of results is listed in Ta-ble 4. The final stellar parameters are shown in the ( T eff ,log g )-diagram with color-coded metallicity (Figure 10).Also we plotted stellar evolutionary tracks in the back-ground (Girardi et al. 2000) with masses between 0.7and 1.9 M (cid:12) and the initial metallicity Z ini =0.019. Ma-jority of the investigated stars have masses between 0.9and 1.5 M (cid:12) . Distributions of the determined T eff , log g ,and [Fe/H] are shown in Figure 11.The determined T eff span from 4700 K to 6950 K (Fig-ure 11a) with a peak at 6100 K, log g range from 3.5 to4.7 (Figure 11b) peaking at 4.3, and [Fe/H] are from − . f e h catalogue of the SIM-BAD database and the PASTEL catalog (Soubiran et al.2010) we found spectroscopic parameters for 39 commonstars (27% of our sample). It was not possible to finda significant number of atmospheric parameters from asingle uniform spectroscopic study. Thus we selectedthe T eff , log g , and [Fe/H] spectroscopic determinationswhich are relatively recent: Galeev et al. (2004); DelgadoMena et al. (2015); Mishenina et al. (2012, 2013); Guill-out et al. (2009); Gonzalez et al. (2010); Takeda et al.(2007); Takeda (2007); Prugniel et al. (2011); Ram´ırezet al. (2013); Fuhrmann (2008); Feltzing & Gustafsson(1998); Fuhrmann (2004); Chen et al. (2000); Gray et al. pectroscopy of dwarf stars around the north celestial pole Table 4.
Sample table of the atmospehric parameters of the programme stars.Tycho2 ID T eff σ T eff log g σ log g [Fe / H] σ [Fe / H] v t σ v t K K km s − km s − TYC 4141-1496-1 6353 54 4.24 0.35 0.01 0.08 1.39 0.24TYC 4141-589-1 6711 93 4.18 0.36 − .
10 0.12 1.76 0.34TYC 4160-145-1 6042 31 3.77 0.22 0.10 0.08 1.33 0.24Notes. This is a small part of the table (the full table is available at the CDS). (2003); Valenti & Fischer (2005). This sample of 39 starswe call the spectroscopic comparison sample.Figure 12 shows a comparison between T eff , log g ,[Fe/H] values of our sample and the spectroscopiccomparison sample of 39 stars (red circles). Anotherso called photometric comparison sample of 119 stars(black dots) was selected from the study by Casagrandeet al. (2011). The effective temperatures in this workare based on the infrared flux method and should bequite accurate.Effective temperature consistency between our sampleand the spectroscopic comparison sample of 39 stars isquite good ( (cid:104) ∆ T eff (cid:105) =14 ±
73 K). The photometric com-parison sample is consistent with our results as well( (cid:104) ∆ T eff (cid:105) = 14 ±
125 K). The identical mean differencebetween photometric and spectroscopic determinations, T eff = 13 ±
95 K, was found by Casagrande et al. (2011)using their sample of 1522 stars.Surface gravities are consistent with both photometric( (cid:104) ∆log g (cid:105) = 0 . ± (cid:104) ∆log g (cid:105) =0 . ± .
09) comparison samples.Metallicities show the same negligible bias ( (cid:104) ∆[Fe / H] (cid:105) = − . ± .
09 dex) for both spectroscopic and photomet-ric comparison samples. The metallicity distribution inour sample of stars meets the one typically observedin the solar neigborhood. For example, Allende Prietoet al. (2004) found (cid:104) [Fe / H] (cid:105) = − . ± .
18 which isidentical to our value (cid:104) [Fe / H] (cid:105) = − . ± . Kinematic parameters
In Figure 13 we present the kinematical distribu-tion of sample stars. According to the distribution ofstars in the Toomre diagramme (the plane a in Fig-ure 13), our sample stars are well confined inside v tot =100 km s − , however, most of them are clearly inside v tot = 50 km s − ( v tot = ( U LSR + V LSR + W LSR ) / ).Currently all these stars are well confined close (dis-tances up to 0.366 kpc) to the Sun (see plane b inFigure 13). In the R mean vs. z max plane our sam-ple stars have mean galactocentric distances from 6to 10 kpc and vertical distances up to 1.3 kpc (the plane c in Figure 13). The plane d in Figure 13 showsthe thick-to-thin disk probability ratios (TD/D) versusmetallicity, computed the same way as in Bensby et al.(2003); Bensby et al. (2014), who show that stars withTD/D > < < TD/D < R mean < R mean > < R gc < c in Fig-ure 13). The velocity dispersions of the sample stars are σ U = 32 km s − , σ V = 21 km s − and σ W = 14 km s − ,that are very close to the characteristic velocity disper-sions (see Bensby et al. 2014) of the thin disk. All of oursample stars are confined inside the v tot = 100 km s − according to the Toomre diagramme, and only two ofour sample stars have TD/D higher than 0.5, thus wecan expect that our sample of stars is a characteristicthin disk sample. The contamination from the thick diskshould be very small.5.3. Spatial metallicity distribution of sample stars
In Figure 14 we present a metallicty distribution inthe Galactic disk according to R mean and | z max | wherethe solid line is a fit to the data with the ± σ confidencebands shown as the shaded ares. A slope of the fit ofthe radial distribution is − . ± .
024 dex kpc − , thusthe radial metallicity gradient of our data is significantby about 2.5 σ . The vertical distribution is − . ± .
099 dex kpc − , thus the vertical metallicity gradientis slightly negative, but significance of the slope is low,only 1 σ .The radial chemical gradients of the thin disk havebeen studied in a number of papers during the recentyears that have used different selection criteria to purifya thin disk population from thick disk memebers. TheRAVE survey (Co¸skunoˇglu et al. 2012, Bilir et al. 2012)0 Mikolaitis et al. N u m be r o f s t a r s T eff , K log g -0.6 -0.4 -0.2 0.0 0.2 0.4 [Fe/H] Figure 11.
Histograms of the determined spectroscopic parameters ( T eff , log g , and [Fe/H]) for all stars in our sample. T e ff , K C O M PA R I S O N � 〈 Δ ������ 〉 �� ����������� 〈 Δ ������ 〉 �� ����������� 〈 Δ ��� g 〉 �� ���������� 〈 Δ ��� g 〉 �� ���������� 〈 Δ T ��� � 〉 �� ������� �� � 〈 Δ T ��� � 〉 �� ����� � Δ T e ff , K T ��� �� ����� ����� l og g C O M PA R I S O N Δ l og g ��� g ����� ����� [ F e / H ] C O M PA R I S O N Δ [ F e / H ] ������ ����� ����� Figure 12.
Comparison between T eff , log g , and [Fe/H] values of our study with values from the spectroscopic comparisonsample (39 stars, red circles) and with values from the photometric comparison sample (119 stars, black dots). The red dashedlines with a slope of 1 are shown for comparison. See the text for more information. provided metallicity gradients separately for giant anddwarf stars for most probable thin-disk stars and found − . ± .
007 dex kpc − . Boeche et al. (2013) foundthe metallicity gradient to be − . ± .
003 dex kpc − for the most probable thin disk sample. A very similarresult was also found in the SEGUE survey (Cheng etal. 2012), a slope of − . ± .
037 dex kpc − for theirsample which was dominated by the thin-disk stars. TheGaia-ESO survey delivered similar gradients in a num-ber of papers: − . ± .
009 dex kpc − by Mikolaitiset al. (2014), − . ± .
008 dex kpc − by Recio-Blancoet al. (2014) and − . ± .
016 dex kpc − by Berge-mann et al. (2014). Anders et al. (2014) in their studyof the APOGEE survey found radial gradients for their HQ and Gold samples to be − . ± .
006 dex kpc − and − . ± .
01 dex kpc − , respectively. The sameauthors repeat similar gradients for samples dominatedby thin disk stars in Anders et al. (2017). Also resultsfrom cepheids should be mentioned. Cepheid variablesare younger than 200 Myr and are clearly thin disk stars.Their radial gradients span from − .
029 dex kpc − (Andrievsky et al. 2002) to − . ± .
003 dex kpc − (Lemasle et al. 2008; Pedicelli et al. 2009).From all this information we see that differences inmetallicity galactocentric gradients can be significant.The reason could be in strategy of building a sam-ple. The results of our sample mostly agree within er-rors with Bergemann et al. (2014), Cheng et al. (2012), pectroscopy of dwarf stars around the north celestial pole
100 50 0 50 V LSR , km s ( U L S R + W L S R ) / , k m s a) 7.95 8.00 8.05 8.10 8.15 8.20 R , kpc0.000.050.100.150.200.250.30 | z | , k p c b) 6 7 8 9 10 R mean , kpc0.00.20.40.60.81.01.21.4 | z m a x | , k p c c) 0.75 0.50 0.25 0.00 0.25[Fe/H]10 T D / D d) Figure 13.
Kinematic parameters of our sample stars. a) Toomre diagram of our sample stars, lines show constant values ofthe total space velocity ( v tot = ( U LSR + V LSR + W LSR ) / ) at 50 and 100 km s − , b) distribution of sample stars in the | z | planevs. R gc , c) distribution of sample stars in z max vs. R mean plane where two vertical dashed lines delimit the solar neighborhood7 < R gc < R mean , kpc1.00.50.00.51.0 [ F e / H ] a) slope= 0.066 ± 0.0240.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4| z max |, kpcb) slope= 0.102 ± 0.099 Figure 14.
Metalicities of the sample stars as a function of the mean galactocentric radii and b) vertical distribution. Thesolid lines show linear fits to the data while shaded areas marks 1 σ uncertainties of the fits. Boeche et al. (2013), and Anders et al. (2014). How-ever, we note that all above-mentioned studies, exceptAnders et al. (2014), computed the gradients accord-ing to a current stellar position. Anders et al. (2014)gradients are computed according to a median Galacto-centric distance, specifically attributing their study tostars that compose the current Solar neighborhood, thustheir results should be more clearly comparable to ours.In general, the chemical gradients of the Galactic diskwith respect to R mean or | z max | have been rarely studiedbefore.The vertical metallicity gradient derived by Bilir et al.(2012) in the thin disk is − . ± .
008 dex kpc − .Boeche et al. (2013) observed a slow metallicity decreasein the range 0 . < | z max | < .
0, where the thin diskis dominant. According to the results by Duong et al.(2018), the thin disk exhibits a steep negative verticalmetallicity gradient, − . ± .
01 dex kpc − . Our re-sults are in broad agreement with the abovementionedstudies, however, we note, that gradients of Bilir et al.(2012) and Duong et al. (2018) are computed accord- ing to current stellar positions while ours is accordingto z max .The formation of the Galactic disk is beyond the mainscope of this study. But we must mention that even if westudy a small sample of stars that are located to quitea narrow direction of the Galaxy, we still can sense theGalactic disk formation history. The observed radialand vertical stellar abundance distributions are resultsof the Galactic disk formation processes and can be com-pared with thin disk evolutionary models. Our elemen-tal abundance radial gradients agree with the modelsof Cescutti et al. (2007), which assumed an inside-outbuild-up of the disk on a time-scale of 7 Gyr in the solarneighborhood (see Matteucci & Francois (1989); Ces-cutti et al. (2007); Pilkington et al. (2012)). On theother hand, it is interesting to note that Haywood et al.(2013) advocated an outside-in formation of the diskthat also should create the negative metallicity gradi-ents in the thin disk. Our results broadly agree with themodel by Toyouchi & Chiba (2018), who studied the gasinfall, re-accretion of out-flowing gas, and radial migra-2 Mikolaitis et al. tion of disk stars, as well as with the thin-disk chemicalevolution models by Minchev et al. (2013, 2014), whichmodelled an inside-out disk formation including stellarmigration, triggered by mergers in the early epochs andafterwards by secular processes of the Galactic bar andspiral arms. SUMMARYThis paper is the first data release of the Spectroscopicand Photometric Survey of the Northern Sky (SPFOT)that aims to provide a detailed chemical compositionfrom high-resolution spectra and photometric variabil-ity data for bright stars in the northern sky using tele-scopes of the Mol˙etai Astronomical Observatory, VilniusUniversity.We have observed high-resolution spectra for all 213photometrically selected 6500 K and cooler dwarfs withmagnitudes up to V = 8 mag in the field with radiusof 20 degrees towards the northern celestial pole. Thisregion of the sky is very important since it will be in-tensively studied by the NASA TESS mission (Rickeret al. 2015; Sullivan et al. 2015) which is an importanttest-bench for the PLATO 2.0 space mission.We determined spectroscopic atmospheric parametersfor 140 slowly rotating stars (for 73% of the stars spec-troscopic parameters were determined for the first time).Our results have no systematic differences when com-pared with other recent studies. Other 73 stars, of the sample of 213, were either fast rotators or double-linebinaries and will be analyzed later using different tech-niques.We also determined a number of kinematic parame-ters that confirmed that almost all of the investigatedstars belong to the kinematic thin disk. We employedthe mean radial and maximal vertical distances and con-firmed that metallicity spatial distributions of brightdwarfs located towards the northern celestial pole com-ply with the latest inside-out thin disk formation mod-els, including those with migration taken into account.Having in mind that only 30% of bright ( V <
Kepler and K Adibekyan V. Z., Sousa S. G., Santos N. C., Delgado MenaE., Gonz´alez Hern´andez J. I., Israelian G., Mayor M.,Khachatryan G., 2012, A&A, 545, A32Allende Prieto C., Barklem P. S., Lambert D. L., CunhaK., 2004, A&A, 420, 183Alonso A., Arribas S., Martinez-Roger C., 1996, A&A, 313,873Alvarez R., Plez B., 1998, A&A, 330, 1109Andrievsky, S. M., Kovtyukh, V. V., Luck, R. E., et al.2002, A&A, 381, 32Anders, F., Chiappini, C., Santiago, B. X., et al. 2014,A&A, 564, A115Anders, F., Chiappini, C., Minchev, I., et al. 2017, A&A,600, A70Auri`ere, M. 2003, EAS Publications Series, 9, 105Baranne A., Mayor M., Poncet J. L., 1979, Vistas inAstronomy, 23, 279Bergemann, M., Ruchti, G. R., Serenelli, A., et al. 2014,A&A, 565, A89 Bergstr¨om H., Peng W. X., Persson A., 1989, Zeitschrift furPhysik D Atoms Molecules Clusters, 13, 203Bensby, T., Feltzing, S., & Lundstr¨om, I. 2003, A&A, 410,527Bensby, T., Feltzing, S., & Oey, M. S. 2014, A&A, 562, A71Bilir, S., Karaali, S., Ak, S., et al. 2012, MNRAS, 421, 3362Blackwell-Whitehead R. J., Pickering J. C., Pearse O.,Nave G., 2005, ApJS, 157, 402Boeche, C., Siebert, A., Piffl, T., et al. 2013, A&A, 559, A59Bovy, J., Allende Prieto, C., Beers, T. C., et al. 2012, ApJ,759, 131Bovy J., 2015, ApJS, 216, 29Brodzinski T., Kronfeldt H.-D., Kropp J.-R., Winkler R.,1987, Zeitschrift fur Physik D Atoms Molecules Clusters,7, 161Brooke J. S. A., Bernath P. F., Schmidt T. W., BacskayG. B., 2013, JQSRT, 124, 11Casagrande L., Sch¨onrich R., Asplund M., Cassisi S.,Ram´ırez I., Mel´endez J., Bensby T., Feltzing S., 2011,A&A, 530, A138 pectroscopy of dwarf stars around the north celestial pole Carigi, L. 2015, Solar Neighborhood, ed. M. Gargaud,W. M. Irvine, R. Amils, H. J. J. Cleaves, D. L. Pinti,J. C. Quintanilla, D. Rouan, T. Spohn, S. Tirard, &M. Viso (Berlin, Heidelberg: Springer Berlin Heidelberg),2286–2287Chen Y. Q., Nissen P. E., Zhao G., Zhang H. W., BenoniT., 2000, A&AS, 141, 491Chen Y. Q., Nissen P. E., Zhao G., 2004, A&A, 425, 697Cheng, J. Y., Rockosi, C. M., Morrison, H. L., et al. 2012,ApJ, 746, 149Co¸skunoˇglu, B., Ak, S., Bilir, S., et al. 2012, MNRAS, 419,2844Cescutti, G., Matteucci, F., Fran¸cois, P., & Chiappini, C.2007, A&A, 462, 943Davis S. J., Wright J. J., Balling L. C., 1971, PhRvA, 3,1220De Pascale M., Worley C. C., de Laverny P., Recio-BlancoA., Hill V., Bijaoui A., 2014, A&A, 570, A68Delgado Mena E., et al., 2015, A&A, 576, A69Dembczy´nski J., Ertmer W., Johann U., Penselin S.,Stinner P., 1979, Zeitschrift fur Physik A Hadrons andNuclei, 291, 207Dorn C., Khan A., Heng K., Connolly J. A. D., Alibert Y.,Benz W., Tackley P., 2015, A&A, 577, A83Dorn C., Hinkel N. R., Venturini J., 2017, A&A, 597, A38Doyle A. P., Davies G. R., Smalley B., Chaplin W. J.,Elsworth Y., 2014, MNRAS, 444, 3592Dulick M., Bauschlicher Jr. C. W., Burrows A., SharpC. M., Ram R. S., Bernath P., 2003, ApJ, 594, 651Duong, L., Freeman, K. C., Asplund, M., et al. 2018,MNRAS,Feltzing S., Gustafsson B., 1998, A&AS, 129, 237Fischer W., H¨uhnermann H., Kollath K.-J., 1967,Zeitschrift fur Physik, 200, 158Fitzpatrick E. L., 1999, PASP, 111, 63Fuhrmann K., 2004, Astronomische Nachrichten, 325, 3Fuhrmann K., 2008, MNRAS, 384, 173Galeev A. I., Bikmaev I. F., Musaev F. A., GalazutdinovG. A., 2004, Astronomy Reports, 48, 492Gilmore G., et al., 2012, The Messenger, 147, 25Girardi L., Bressan A., Bertelli G., Chiosi C., 2000, A&AS,141, 371Gonzalez G., Carlson M. K., Tobin R. W., 2010, MNRAS,403, 1368Gray R. O., Corbally C. J., Garrison R. F., McFaddenM. T., Robinson P. E., 2003, AJ, 126, 2048Guillout P., et al., 2009, A&A, 504, 829Gustafsson B., Edvardsson B., Eriksson K., JørgensenU. G., Nordlund ˚A., Plez B., 2008, A&A, 486, 951 Hakkila J., Myers J. M., Stidham B. J., Hartmann D. H.,1997, AJ, 114, 2043Haywood, M., Di Matteo, P., Lehnert, M. D., Katz, D., &G´omez, A. 2013, A&A, 560, A109Handrich E., Steudel A., Walther H., 1969, Physics LettersA, 29, 486Heiter U., et al., 2015, PhyS, 90, 054010Heiter, U., Jofr´e, P., Gustafsson, B., et al. 2015, A&A, 582,A49Hermann G., Lasnitschka G., Schwabe C., Spengler D.,1993, Spectrochimica Acta, 48, 1259Wallace, L., Hinkle, K. H., Livingston, W. C., & Davis,S. P. 2011, ApJS, 195, 6Høg E., et al., 2000, A&A, 355, L27Jofr´e, P., Heiter, U., Soubiran, C., et al. 2014, A&A, 564,A133Johann U., Dembczy´nski J., Ertmer W., 1981, Zeitschriftfur Physik A Hadrons and Nuclei, 303, 7Joshi Y. C., 2007, MNRAS, 378, 768Jurgenson C. A., Fischer D. A., McCracken T. M., StollR. A., Szymkowiak A. E., Giguere M. J., Santoro F. G.,Muller G., 2014, in Ground-based and AirborneInstrumentation for Astronomy V. p. 91477F,doi:10.1117/12.2056491Jurgenson C., Fischer D., McCracken T., Sawyer D.,Giguere M., Szymkowiak A., Santoro F., Muller G., 2016,Journal of Astronomical Instrumentation, 5, 1650003Kaufer, A., Stahl, O., Tubbesing, S., et al. 1999, TheMessenger, 95, 8Kawata, D., Bovy, J., Matsunaga, N., & Baba, J. 2018,arXiv:1803.05927Kerr, F. J., & Lynden-Bell, D. 1986, MNRAS, 221, 1023Kurucz R., 1993, Diatomic Molecular Data for OpacityCalculations. Kurucz CD-ROM No. 15. Cambridge,Mass.: Smithsonian Astrophysical Observatory, 1993., 15Lemasle, B., Fran¸cois, P., Piersimoni, A., et al. 2008, A&A,490, 613Majewski S. R., et al., 2017, AJ, 154, 94Masseron T., et al., 2014, A&A, 571, A47Matteucci, F. & Francois, P. 1989, MNRAS, 239, 885Mayor M., 1985, in Philip A. G. D., Latham D. W., eds,Vol. 88, Stellar Radial Velocities. pp 35–48Mayor, M., Pepe, F., Queloz, D., et al. 2003, TheMessenger, 114, 20Michalik D., Lindegren L., Hobbs D., 2015, A&A, 574, A115Miglio A., et al., 2017, Astronomische Nachrichten, 338, 644Mikolaitis ˇS., et al., 2014, A&A, 572, A33Minchev, I., Chiappini, C., & Martig, M. 2013, A&A, 558,A9 Mikolaitis et al.
Minchev, I., Chiappini, C., & Martig, M. 2014, A&A, 572,A92Mishenina T. V., Soubiran C., Kovtyukh V. V., KatsovaM. M., Livshits M. A., 2012, A&A, 547, A106Mishenina T. V., Pignatari M., Korotin S. A., Soubiran C.,Charbonnel C., Thielemann F.-K., Gorbaneva T. I.,Basak N. Y., 2013, A&A, 552, A128Mishenina T., Gorbaneva T., Pignatari M., ThielemannF.-K., Korotin S. A., 2015, MNRAS, 454, 1585Mitchell Jr. W. E., Mohler O. C., 1965, ApJ, 141, 1126Molenda- ˙Zakowicz, J., Frasca, A., De Cat, P., & Catanzaro,G. 2017, in European Physical Journal Web ofConferences, Vol. 152, European Physical Journal Web ofConferences, 04002Nascimbeni V., et al., 2016, MNRAS, 463, 4210Nelder J. A., Mead R., 1965, Computer Journal, 7, 308Niemczura, E., Poli´nska, M., Murphy, S. J., et al. 2017,MNRAS, 470, 2870Nordstr¨om B., et al., 2004, A&A, 418, 989Perryman M. A. C., et al., 1997, A&A, 323, L49Petigura, E. A., Howard, A. W., Marcy, G. W., et al. 2017,AJ, 154, 107Pedicelli, S., Bono, G., Lemasle, B., et al. 2009, A&A, 504,81Pilkington, K., Few, C. G., Gibson, B. K., et al. 2012,A&A, 540, A56Pourbaix D., et al., 2004, A&A, 424, 727Prugniel P., Vauglin I., Koleva M., 2011, A&A, 531, A165Prˇsa, A., Harmanec, P., Torres, G., et al. 2016, AJ, 152, 41Ram R. S., Brooke J. S. A., Bernath P. F., Sneden C.,Lucatello S., 2014, ApJS, 211, 5Ram´ırez I., Allende Prieto C., Lambert D. L., 2013, ApJ,764, 78Rauer H., et al., 2014, Experimental Astronomy, 38, 249Rauer H., Aerts C., Cabrera J., PLATO Team 2016,Astronomische Nachrichten, 337, 961Recio-Blanco, A., de Laverny, P., Kordopatis, G., et al.2014, A&A, 567, A5Reid, M. J., Menten, K. M., Zheng, X. W., et al. 2009,ApJ, 700, 137-148Reid, M. J., & Dame, T. M. 2016, ApJ, 832, 159 Ricker G. R., et al., 2014, in Space Telescopes andInstrumentation 2014: Optical, Infrared, and MillimeterWave. p. 914320 ( arXiv:1406.0151arXiv:1406.0151