GAUFRE: a tool for an automated determination of atmospheric parameters from spectroscopy

0.22 MB·English

by  Marica Valentini

GAUFRE: a tool for an automated determination of
atmospheric parameters from spectroscopy
M.Valentini1,a,T.Morel1,A.Miglio2,L.Fossati3,andU.Munari4
1 Instituted’AstrophysiqueetdeGe´ophysique,Universite´ deLie´ge,B-4000Lie´ge,Belgium
2 SchoolofPhysicsandAstronomy,UniversityofBirmingham,UnitedKingdom
3 Argelander-Institut fu¨r Astronomie der Universita¨t Bonn, Auf dem Hu¨gel 71, 53121, Bonn, Ger-
3 many
1 4 INAF-OAPd,OsservatorioAstronomicodiPadova,Padova,Italy
0
2

n Abstract. We present an automated tool for measuring atmospheric parameters (Teff,
a log g), [Fe/H]) for F-G-K dwarf and giant stars. The tool, called GAUFRE, is written
J inC++andcomposedofseveralroutines:GAUFRE-RVmeasuresradialvelocityfrom

0 spectraviacross-correlationagainstasynthetictemplate,GAUFRE-EWmeasuresatmo-
3 spheric parameters through the classic line-by-line technique and GAUFRE-CHI2 per-
formsaχ2fittingtoalibraryofsyntheticspectra.AsetofF-G-Kstarsextensivelystudied
]
R intheliteraturewereusedasabenchmarkfortheprogram:theirhighsignal-to-noiseand
highresolutionspectrawereanalyzedbyusingGAUFREandresultswerecomparedwith
S
thosepresentinliterature.Thetoolisalsoimplementedinordertoperformthespectral
.
h analysisafterfixingthesurfacegravity(logg)totheaccuratevalueprovidedbyastero-
p seismology.AsetofCoRoTstars,belongingtoLRc01andLRa01fieldswasusedforfirst
- testingtheperformancesandthebehavioroftheprogramwhenusingtheseismiclogg.
o
r
t
s 1 Introduction
a
[

Spectroscopyisoneofthemostpowerfultoolthatastronomypossessesinordertoderiveatmospheric
1 parametersandabundancesofstars.Fromthestellarspectrumitispossibletomeasuretheeffective
v
6 temperature(Teff),thesurfacegravity(logg)andtheabundanceofiron([Fe/H])andofseveralother
5 elements.
2 The classic method consists in measuring the equivalent widths (EW) of a species in two different
7 ionization states, usually FeI and FeII. By imposing excitation and ionization equilibrium through
1. stellaratmospheremodels,itispossibletoderiveTeff,loggandtoinferelementalabundancesfrom
0 thecurvesofgrowth.Thismethodispreciseanditiswidelyadopted[1].Inspiteoftheprecision,the
3 line-by-lineclassicalanalysisistimeconsuming:usuallyEWsaremeasuredbyhandusingtheIRAF
1 “splot”routine(wherethechoiceofthecontinuumandthepositionofthelineiscompletelymanual)
v: andtheprocedureforfindingthebestTeff andloggrequiresalargenumberofiterations.
The growing amount of stellar data due to large surveys (i.e. RAVE, SEGUE, LAMOST, HERMES
i
X and Gaia ESO) and the availability of dedicated telescopes and multi-object spectrograph, requires
r thedevelopmentofautomatedpipelines,methodsandprogramsthatfastentheanalysisprocess.For
a
example, DAOSPEC [2] and ARES [3] are useful and free codes that performs an automated EW
measurement;theMOOGcode[4]isavalidcollectionofroutinesusefulfordeterminingatmospheric
parametersandabundances.Inparticular,MOOGabfindandsynthtasksarewidelyusedinliterature
[5][1].
Anothermethodfortheestimationofatmosphericparametersandelementalabundancesistocompute
a e-mail:[e-mail]
EPJWebofConferences
Table1.ComparisonbetweentheradialvelocitywemeasuredwithGAUFRE-RV(adoptingthelibraryofsyn-
theticspectraofFossati)andIAUvalues.Allradialvelocitiesareinkms−1
Star Instrument IAU GAUFRE-RV
v σ v σ
rad vrad rad vrad
HD003712 Asiago-Echelle −3.9 0.1 −4.1 0.3
HD012929 Asiago-Echelle −14.3 0.2 −13.8 0.4
ESO-FEROS −14.6 0.2
HD062509 Asiago-Echelle +3.3 0.1 +3.3 0.6
ESO-UVES +3.4 0.3
HD065934 Asiago-Echelle +35.0 0.3 +34.6 0.5
HD090861 Asiago-Echelle +36.3 0.4 +36.7 0.4
HD212943 Asiago-Echelle +54.3 0.3 +54.2 0.4
ESO-UVES +54.1 0.2
HD213014 Asiago-Echelle −39.7 0.0 −40.0 0.5
asetofsyntheticspectraandtofindthebestmatchbetweenthesyntheticandtheobservedspectrum.
ThismethodisadoptedbysurveysasRAVE[6]andHERMES[7]aswellasinsmallsurveysasARCS
[8][9].
Inthispaperwepresentanewautomaticcode,GAUFRE,thatcanperformbothtypeofanalysis.In
section2wegiveashortdescriptionoftheideabehindtheprogramandthetwotypesofanalysisthat
itperforms.Insubsection2.4wepresentthefirstresultsoftestsusingspectraofobjectswellknown
inliterature.Insection3wepresentanadditionaltoolofGAUFREthatconsistsinusingtheastero-
seismicgravityasafixedvalueforlog(g)inordertorefinethemeasurementofTeff,microturbulence
velocity(ξ )andelementalabundances.Section4discussesthefutureperspectivesforthecode.
mic
2 The GAUFRE code
GAUFREisacollectionofseveralC++routines.Itiswritteninordertomeasurefastandprecisely
radial velocity (Vrad) and atmospheric parameters (Teff, log g, [Fe/H]) of a star starting from a one-
dimensionalnormalizedspectrum.Theradialvelocityismeasuredwithacross-correlationtechnique
(routineGAUFRE-RV),atmosphericparameterscanbemeasuredadoptingaχ2 fittingoveralibrary
ofsyntheticspectra(GAUFRE-CHI2routine)orwiththeclassicFeI-FeIIlinestechnique(GAUFRE-
EWroutine).
GAUFREisanupdatedandextendedversionofthecodewrittenfortheAsiagoRedClumpSpectro-
scopicsurvey(ARCS)[8].GAUFREwascreatedbecauseoftheneedofintroducingnewlibrariesof
syntheticspectra,adaptingthecodefordifferentresolutionsandimplementingtheclassicline-by-line
technique. The code was written in C++ and does not need any particular library, in order to avoid
any software license problems and to be executable in different platforms. The only additional pro-
gram needed for running GAUFRE is MOOG, which can be easily installed and downloaded from
its homepage (http://www.as.utexas.edu/∼chris/moog.html). The user is also supposed to download
the libraries of synthetic spectra and model atmospheres required by GAUFRE. So far GAUFRE is
tested for F-G-K stars (both giants and dwarfs) and works in the 3500 - 9000 Å spectral range.
The program has currently been tested for spectral resolutions of the Asiago Echelle Spectrograph
(R=20,000), ESO-FLAMES-GIRAFFE HR10, HR15N, HR21, HR14 setups (R=16,000 - 20,000),
ESO-FLAMES-UVESU580setup(R=52,000),ESO-FEROS(R=48,000)andAAOmega2dF580V
and385R(R=1,300).Detailsabouttheadoptedlibrariescanbefoundonsubsections2.3and2.4.
ConferenceTitle,tobefilled
Table2.LibrariesofsyntheticspectrausedbyGAUFRE-RVandGAUFRE-CHI2.
Library Model Linelist Solarabundances Resolution
atmospheres
Munarietal.(2005) Castelli&Kurucz(2003) Kurucz Grevesse&Sauval(1998) 20,000
Fossati Castelli&Kurucz(2003) VALD2 Grevesse&Sauval(1998) 1,000,000
DeLavernyetal.(2012) MARCS Grevesse,Asplund&Sauval(2007) 150,000
2.1 RadialVelocities:GAUFRE-RV
Theradialvelocitysubroutine(GAUFRE-RV)measurestheradialvelocityV bycross-correlating
rad
theobservedspectrumwithasyntheticspectrallibrary[10].Theprocedureisthesameasdescribed
in [8]. The procedure starts from the continuum normalized spectrum in a 2-column ASCII format.
First, the synthetic normalized spectra of the selected library are renormalized, following the same
parametersasadoptedforthenormalizationoftheobservedspectrum(samefunction,sameorderand
samehighandlowrejectionvalues).Tolowertheimpactofthedifferentnoiselevelintheobserved
andsyntheticspectra,theyarescaledtomatchtheirgeometricmean.Thisprocedureisneededinor-
der to improve the accuracy of the cross-correlation and of the χ2 fitting and it is performed by the
GAUFRE-LIBroutine(seealsosection2.2).
TheresultoftheGAUFRE-RVsubroutineisafilecontainingthevalueofV andanasciifilecon-
rad
tainingthecontinuumnormalizedspectrumcorrectedfortheradialvelocity.
TovalidatetheGAUFRE-RVroutinewemeasuredtheradialvelocityofasetofRedGiantsstarsthat
are IAU standard radial velocity stars. These stars were observed with the Asiago Echelle Spectro-
graph(INAF-OAPd);wethendownloaded,whenavailable,thespectrumfromtheESO-Archive.For
thistest,weadoptedthesyntheticspectralibraryprovidedbyL.Fossati(cf.section2.2).Resultsare
summarizedinTable1.Themeandifferencebetweenourvaluesandthosepresentintheliteratureis
∆RV =0.10kms−1,witharmsof0.3kms−1.
(cid:12)
2.2 χ2onsyntheticspectralibraries:GAUFRE-CHI2
Atmosphericparameters(Teff,logg,[Fe/H],[α/Fe],Vrotsin i)areobtainedbyGAUFRE-CHI2viaχ2
fittingofthecontinuumnormalizedspectrumagainstasyntheticspectrallibrary.
Thechoiceofthespectrallibraryisuptotheuser,availablelibrariesare:alibrarybasedonKurucz
modelatmosperes[11],thelibraryprovidedbyL.Fossati(builtadoptingthespectralsynthesiscode
Synth3describedin[12])andtheAMBRElibrary[13].Characteristicsofdifferentlibrariesaresum-
marizedinTable3.Librariesareprovidedatdifferentresolutionsandtheycoverawidespectralrange
(usually 3,500 - 10,000 Å). Before starting χ2 analysis, the desired library must be cut, normalized
anddegradedatthesamewavelengthinterval,normalizationfunctionandresolutionoftherealspec-
tra.ForthispurposearoutineGAUFRE-LIBhasbeencreated.Thedegradationofthespectraatthe
desiredresolutionisperformedthroughdeconvolutionwithaGaussianprofile.
2.3 EWandMOOG:GAUFRE-EW
GAUFRE-EWautomaticallyperformstheclassicalline-by-lineanalysisforderivingatmosphericpa-
rameters and abundances. The procedure starts from a continuum normalized spectrum (ASCII 2
columns format), a file containing a list of lines to measure and their parameters (as requested by
MOOG)andafilecontainingtheparametersofthespectrumlikethewavelengthcoverage,resolution
and,ifany,guessedvaluesforTeff andlogg.
The EW of every line present in the input file is measured (except when it is not detectable). The
EPJWebofConferences
300
250 200
RE−Lit. 1000
AUF −100
200 G
−200
−300
3500 4000 4500 5000 5500 6000 6500
1
150 RE−Lit. 0.05 ECWHI2
UF
A
G −0.5
100
−1
0 1 2 3 4 5 6
0.6
0.4
50 RE−Lit. 0.02
AUF −0.2
G
−0.4
−0.6
50 100 150 200 250
−1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6
Literature
Fig.1.Leftpanel:ComparisonbetweentheEWmeasuredbyGAUFRE-EW(y-axis)andtheEWmeasuredwith
theIRAF-splotroutine,fittingthelinetoaGaussianprofile.Rightpanel:DifferencesbetweenthevaluesofTeff
(toppanel),logg(middlepanel)and[Fe/H](bottompanel)measuredbyGAUFE-EWandGAUFRE-CHI2and
thosepresentinliterature.
program selects an area of 3-4 Å around the wavelength of the line (this parameter is selected by
theuser).Thespectrumisthenfittedwithapolynomialfunctioninordertodeterminethecontinuum
and point with the lowest intensity. In order to test the automated EW measurement of the line we
comparedtheEWvaluesobtainedbymeasuringfeaturesbyhand(usingIRAF-splot)andthecorre-
spondentEWmeasuredbyGAUFRE.ThetestwasperformedonthespectrumofArcturus,takenwith
the ESO-FLAMES-UVES instrument, with the U580 setup (5770-6825 Å). The agreement is quite
goodandrmsisof∼3mÅ(seeFigure1).
Atmospheric parameters were computed by using MOOG abfind driver (the program uses its non-
iinteractiveversion,MOOGSILENT),themeasuredEWofFeIandFeIIlinesandafamilyofmodel
atmospheres(MARCS[14]orKurucz[15]).Teff iscalculatedbyassumingtheexcitationequilibrium
andminimizingthetrendoftheFeabundanceversustheexcitationpotential.Thesurfacegravity,logg
isderivedbyassumingtheionizationequilibrium:logn(FeI)=logn(FeII).Theprocedureisiterative
and the program will converge to Teff and log g that satisfy both the ionization and excitation equi-
libria.Thevalueofthemicroturbulenceξ isderivedbyminimizingthetrendoftheFeIabundance
mic
versustheFeIlinesEW.
2.4 Atmosphericparametersvalidation
Wetookspectraof7F-G-KstarstakenwithFLAMES-UVESU580orFEROSfromtheESO-archive
(http://archive.eso.org).Weselectedspectraoftargetsverywellknowninliterature:αCenA,µCasA,
βVir,Arcturus,µLeo,ξ HyaandγSge.Asareferenceweusedanaveragevalueofthemostrecent
entries of the PASTEL catalog [3]. We analyzed spectra with both GAUFRE-EW (Kurucz model
atmospheres)andGAUFRE-CHI2(Fossatilibraryofsyntheticspectra)andwecomparedourresults
withthosepresentinliterature.TheagreementisquitegoodasshowedinTab.3.
ConferenceTitle,tobefilled
Table3.ComparisonoftheatmosphericparametersobtainedbyGAUFRE-CHI2(G-CHI2)andGAUFRE-EW
(G-EW)withthopsepresentinliteratureforasetof7stars.
GAUFRE-CHI2 GAUFRE-EW
(cid:104)Teff|G−CHI2−Teff(cid:105)=16K σ=84K (cid:104)Teff|G−EW−Teff(cid:105)=0K σ=17K
(cid:104)logg| −logg(cid:105)=0.13dex σ=0.17dex (cid:104)logg| −logg(cid:105)=0.11dex σ=0.17dex
G−CHI2 G−EW
(cid:104)[Fe/H]| −[Fe/H](cid:105)=-0.05dex σ=0.10dex (cid:104)[Fe/H]| −[Fe/H](cid:105)=-0.06dex σ=0.10dex
G−CHI2 G−EW
3 Asteroseismic constraints
Asextensivelydiscussed,forexample,in[16],thefrequencyofmaximumpower,ν canbeusedfor
max
derivingapreciseestimateofthesurfacegravity:
(cid:32) (cid:33) (cid:32) (cid:33)
logg=logg +log νmax + 1log Teff . (1)
(cid:12)
νmax,(cid:12) 2 Teff,(cid:12)
TheprecisionofthegravityderivedfromEq.1isgenerallyexpectedtobebelow0.05dex,thanksto
theweaksensitivityofthescalingrelationtotheassumedTeff andthehighprecisionusuallyachieved
forthemeasurementofν .
max
Thescalingrelationhasbeendemonstratedtobereliable[17][18][19][20]anditcanbeusedforre-
finingatmosphericparametersandabundances.Asamatteroffactonecanimprovethespectroscopic
analysisbyfixingtheloggtotheseismicvalue:thislargelyincreasestheaccuracyofthederivedTeff,
[Fe/H]andhencechemicalabundances.
Asetofspectraof111RGstarsbelongingtotheLRc01andLRa01fieldsofCoRoThavebeenusedas
benchmarkfortestingtheloggvaluesderivedbyGAUFRE.SpectraaretakenwiththeESO-FLAMES
GIRAFFE9setup,centeredintheMgItriplet(5278Å).Spectrahavealreadybeenanalyzed[21]by
usingMATISSEtool[22].
First, we compared logg values of GAUFRE-CHI2 (using the synthetic library provided by L. Fos-
sati,seeTab3)withthoseprovidedbyasteroseismology(seeFig2,toppanels),thenwecomparedthe
values of Teff and [Fe/H] derived by fixing logg to the seismic value (see Fig2, middle and bottom
panels).ThesetofspectratakenfromtheworkofGazzanohasverypoorSNR(seeFig2),of∼30on
average.InaddictionthespectralrangecoveredbyspectraisaffectedbythepresenceofMgHmolec-
ularbands.Thiscomplicatesevenmorethecontinuumnormalization,leadingtoseveralsystematics
intheatmosphericparametersdetermination.
As shown in Fig. 2, the seismic gravity can be used to enhance the accuracy of the atmospheric
parameters.InthecaseofGazzanoetal.dataset[21]withapoorSNR,theseismicgravityhelpedin
providing a more precise determination of Teff and [M/H] reducing average errors from 120 K and
0.20dexto75Kand0.11dexrespectively.
4 Conclusions
We present the GAUFRE program, a versatile tool developed for measuring radial velocities and at-
mosphericparametersfromopticalspectra.
Theprogramiscomposedbydifferentroutines:GAUFRE-RV,GAUFRE-CHI2,GAUFRE-EW.
Weperformedsomepreliminarytestsinordertochecktheperformancesofthetool.Thesetestsshow
thattheprogramisreliableandthatitcanbeusedtoprocessspectraofF-G-Kdwarfsandgiantswith
aSNRabove40.AtthemomentitisadoptedbytheLie`genodewithintheGaia-ESOSurvey(PI:G.
GilmoreandS.Randich)andfurtherapplicationsareplanned.
WealsoshowedaninterestingandusefulextensionofGAUFREthatuses,whenavaiable,theseismic
loggasafixedvalueforthesurfacegravity:thepreciseandalsolikelymoreaccuratevaluesgivenby
asteroseismologyallowustogreatlyrefinethevaluesofTeff and[Fe/H].
The GAUFRE tool is continously in development: we plan to implement new libraries of synthetic
EPJWebofConferences
3 3
2 2
∆ log(g) (dex)−011 ∆ log(g) (dex)−011
−2 −2
0 20 40 60 80 100 120 1.5 2 2.5 3 3.5
SNR log(g)SISMO (dex)
500
1000
∆ Teff (K) −5500000 ∆ Teff (K) −5000
−1000 −1000
0 20 40 60 80 100 120 4000 4200 4400 4600 4800 5000 5200
SNR Teff (K)SISMO (K)
1 1
[M/H] (dex)−00−..0155 ∆ [M/H] (dex)−00−..0155
−1.5 −1.5
0 20 40 60 80 100 120 −1 −0.5 0 0.5 1
S/N [M/H]SISMO (dex)
Fig.2.Leftpanel:differencesbetweentheatmosphericparametersmeasuredwithdifferenttechniques(GAUFRE-
CHI2thispaper,MATISSE[21],photometry(J−K)andasteroseismology)infunctionoftheSNR.Rightpanel:
differencesbetweentheatmosphericparametersmeasuredwithdifferenttechniques(GAUFRE-CHI2thispaper,
MATISSEby[21],photometry(J−K)andasteroseismology)infunctionoftheTeff,loggand[M/H]measured
by GAUFRE-CHI2 by fixing the log g to the seismic value. Green crosses are data from [21], blue diamonds
aredataobtainedbyusingGAUFRE-CHI2andredcirclesarevaluesmeasuredbyGAUFRE-CHI2byfixingthe
loggtotheseismicvalue.Itisworthtotakeintoaccountthepoorqualityofthespectra:morethan50%ofthe
spectrapossessaSNR<40.
spectraandtoextendtheanalysistotheinfraredregion.Newanddetailedtestsareplannedaswell,in
ordertobetterinvestigatetheperformancesofGAUFREatdifferentSNR.
InthenextfutureweplantomaketheGAUFREcodeavaiablethroughthewebandtocreateanuser-
friendlygraphicalinterface.
MV aknowledges financial support from Belspo for contract PRODEX COROT. TM acknowledges finan-
cialsupportfromBelspoforcontractPRODEXGAIA-DPAC.PaperbasedonobservationscollectedatAsiago
ObservatoryandfromdatarecoveredfromtheESOArchive.
References
1. T.Morel,A.Miglio,MNRAS419,L34(2012),1110.2036
2. P.B.Stetson,E.Pancino,PASP120,1332(2008),0811.2932
3. S.G. Sousa, ARES: Automatic Routine for line Equivalent widths in stellar Spectra (2012), astrophysics
SourceCodeLibrary,1205.009
4. C. Sneden, J. Bean, I. Ivans, S. Lucatello, J. Sobeck, MOOG: LTE line analysis and spectrum synthesis
(2012),astrophysicsSourceCodeLibrary,1202.009
5. S. Randich, P. Sestito, F. Primas, R. Pallavicini, L. Pasquini, A&A450, 557 (2006),
arXiv:astro-ph/0601239
6. M. Steinmetz, RAVE: the RAdial Velocity Experiment, in GAIA Spectroscopy: Science and Technology,
edited by U. Munari (2003), Vol. 298 of Astronomical Society of the Pacific Conference Series, p. 381,
arXiv:astro-ph/0211417
ConferenceTitle,tobefilled
7. K.C. Freeman, The HERMES Project: Reconstructing Galaxy Formation, in Galactic Archaeology: Near-
FieldCosmologyandtheFormationoftheMilkyWay,editedbyW.Aoki,M.Ishigaki,T.Suda,T.Tsujimoto,
N.Arimoto(2012),Vol.458ofAstronomicalSocietyofthePacificConferenceSeries,p.393
8. M.Valentini,U.Munari,A&A522,A79(2010),1007.0207
9. T.Saguner,U.Munari,M.Fiorucci,A.Vallenari,A&A527,A40(2011),1012.5373
10. J.Tonry,M.Davis,AJ84,1511(1979)
11. U.Munari,R.Sordo,F.Castelli,T.Zwitter,A&A442,1127(2005),arXiv:astro-ph/0502047
12. O.P.Kochukhov, Spectrum synthesis for magnetic, chemicallystratified stellar atmospheres, in Physics of
MagneticStars,editedbyI.I.Romanyuk,D.O.Kudryavtsev,O.M.Neizvestnaya,V.M.Shapoval(2007),pp.
109–118,arXiv:astro-ph/0701084
13. P.deLaverny,A.Recio-Blanco,C.C.Worley,B.Plez,A&A544,A126(2012),1205.2270
14. B.Gustafsson,B.Edvardsson,K.Eriksson,U.G.Jørgensen,Å.Nordlund,B.Plez,A&A486,951(2008),
0805.0554
15. F.Castelli,R.L.Kurucz,ArXivAstrophysicse-prints(2004),arXiv:astro-ph/0405087
16. T.Morel,A.Miglio,N.Lagarde,J.Montalban,M.Rainer,E.Poretti,S.Hekker,T.Kallinger,B.Mosser,
M.Valentinietal.,ArXive-prints(2012),1210.2982
17. H.Bruntt,T.R.Bedding,P.O.Quirion,G.LoCurto,F.Carrier,B.Smalley,T.H.Dall,T.Arentoft,M.Bazot,
R.P.Butler,MNRAS405,1907(2010),1002.4268
18. D. Stello, D. Huber, T. Kallinger, S. Basu, B. Mosser, S. Hekker, S. Mathur, R.A. Garc´ıa, T.R. Bedding,
H.Kjeldsenetal.,ApJ737,L10(2011),1107.0490
19. A.Miglio,J.Montalba´n,A.Noels,RedGiantsasProbesoftheStructureandEvolutionoftheMilkyWay
(2012)
20. D. Huber, M.J. Ireland, T.R. Bedding, I.M. Branda˜o, L. Piau, V. Maestro, T.R. White, H. Bruntt,
L.Casagrande,J.Molenda-Z˙akowiczetal.,ApJ760,32(2012),1210.0012
21. J.C.Gazzano,P.deLaverny,M.Deleuil,A.Recio-Blanco,F.Bouchy,C.Moutou,A.Bijaoui,C.Ordenovic,
D.Gandolfi,B.Loeillet,A&A523,A91(2010),1011.5335
22. A.Recio-Blanco,A.Bijaoui,P.deLaverny,MNRAS370,141(2006),arXiv:astro-ph/0604385